Skip to main content

Full text of "Journal of the Society of Motion Picture Engineers"

See other formats





Volume XXXIV January, 1940 



Report on the Adaptation of Fine-Grain Films to Variable- 
Density Sound Technics 3 

Improvement in Sound and Picture Release Through the Use 

of Fine-Grain Film C. R. DAILY 12 

Photographic Duping of Variable-Area Sound 


A Sound-Track Center-Line Measuring Device 


Starting Characteristics of Speech Sounds 


Volume Distortion S. L. REICHES 59 

Lenses for Amateur Motion Picture Equipment 


Report of the Standards Committee 88 

Report of the Studio Lighting Committee 94 

The Importance of Cooperation between Story Construction 

and Sound to Achieve a New Personality in Pictures 

L. L. RYDER 98 
New Motion Picture Apparatus 

A Multiduty Motor System A. L. HOLCOMB 103 

Current Literature 114 

Book Review 116 

1940 Spring Convention at Atlantic City, N. J., April 22nd-25th, 

Inclusive 117 

Society Announcements 120 





Board of Editors 

J. I. CRABTREE, Chairman ^ 




Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscription or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., East on, Pa., or Hotel Pennsylvania, New York, N. Y. 
Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 

West Coast Office, Suite 226, Equitable Bldg., Hollywood, Calif. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1940, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is not 
responsible for statements made by authors. 


* President: E. A. WILLIFORD, 30 East 42nd St., New York, N. Y. 

* Past-President: S. K. WOLF, RKO Building, New York, N. Y. 

* Executive Vice-President: N. LEVINSON, Burbank, Calif. 

** Engineering Vice-President: D. E. HYNDMAN, 350 Madison Ave., New York, 
N. Y. 

* Editorial Vice-President: J. I. CRABTREE, Kodak Park, Rochester, N. Y. 

** Financial Vice-President : A. S. DICKINSON, 28 W. 44th St., New York, N. Y. 

* Convention Vice-President: W. C. KUNZMANN, Box 6087, Cleveland, Ohio. 

* Secretary: J. FRANK, JR., 356 W. 44th St., New York, N. Y. 

* Treasurer: R. O. STROCK, 35-11 35th St., Astoria, Long Island, N. Y. 


* M. C. BATSEL, Front and Market Sts., Camden, N. J. 

* J. A. DUBRAY, 1801 Larchmont Ave., Chicago, 111. 

** A. N. GOLDSMITH, 444 Madison Ave., New York, N. Y. 
** H. GRIFFIN, 90 Gold St., New York, N. Y. 

* P. J. LARSEN, 29 S. Munn Ave., East Orange, N. J. 

* L. L. RYDER, 5451 Marathon St., Hollywood, Calif. 

** A. C. HARDY, Massachusetts Institute of Technology, Cambridge, Mass. 

* H. G. TASKER, 14065 Valley Vista Blvd., Van Nuys, Calif. 

* Term expires December 31, 1940. 

JL \_ i in ^ A |y 1 1 w j .L-' v\_ \. 1 1 1 ut_ i *j A. f j. cr-xvs . 

** Term expires December 31, 1941. 


Summary. The activities of various West Coast studios and laboratories in the 
adaptation of fine -grain films for variable-density recording are summarized. A set 
of requirements for a suitable variable density negative and print emulsions have been 
formulated and are included in the paper. 

Since the latter part of 1938 and continuing up to the present time, 
considerable experimentation has been carried on by Electrical Re- 
search Products, Inc., and several of the Western Electric recording 
licensees with a view to utilizing some of the fine-grain emulsions for 
variable-density negative and re-recording print use with the expecta- 
tions that such films would be much freer from background noise, 
and at the same time provide better sound quality than that obtain- 
able from standard positive type emulsions. The Metro-Goldwyn- 
Mayer Studio, after a thorough investigation of various fine-grain 
emulsions, have adopted the type 222 stock furnished by the Dupont 
Film Manufacturing Corporation for all original negative and print 
material, as well as for the re-recorded negative. Several pictures 
have been released in which fine-grain film has been used in some 
of the processes leading up to the release print. At the Samuel 
Goldwyn Studios considerable tests were carried out on various ex- 
perimental emulsions offered by Dupont and by the Eastman Kodak 
Company, as well as on some of the standard fine-grain emulsions 
offered by both these suppliers. The recording tests made by this 
studio were confined to original negatives and re-recording prints of 
scoring, and some of this material has been utilized in pictures that 
have been already released. For some time all re-recording prints 
at Paramount have been made on the 222 stock from a normal 
sound negative and one complete picture has been re-recorded to 
fine-grain release negative with a limited number of release movietone 
prints also being printed on the 222 stock. This represents the 
first complete adaptation of fine-grain stocks to release. The Uni- 
versal Studio made some scoring and dialog tests on the 1365 stock 

* Presented at the 1939 Fall Meeting at New York, N. Y. ; received October 
13, 1939. 



furnished by the Eastman Kodak Company, with prints being made 
on the same material. 

Simultaneously with this program in the studios, Electrical Re- 
search Products, Inc., has been active in a fundamental investigation 
of the sensitometric properties and the signal-to-noise relationships 
of all the available fine-grain emulsions, with a view to giving the 
results to its recording licensees for such value as they might have in 
the working out of variable-density recording programs. In order to 
correlate the activities of these diverse groups, a Committee of repre- 
sentatives from the Western Electric Recording licensees, the major 
film suppliers, and Electrical Research Products, Inc., was formed in 
February of this year. The exchange of information between the 
various members participating in these Committee meetings has 
tended to eliminate a lot of duplicate efforts that would otherwise be 
inevitable under such circumstances. The Committee also set 
about to draw up a set of requirements for an ideal fine-grain variable- 
density negative and re-recording print stock. These requirements, 
listed in an Appendix to this report, were submitted to the repre- 
sentatives of the major film suppliers in May of this year. As will be 
noted from the Appendix, the attention of the Committee was con- 
fined to securing films more suitable for those processes which did not 
in any way conflict with quality of the picture as finally seen on the 
screen. Thus, the specifications are confined to original sound nega- 
tive, original or re-recording print, and re-recording negative; a 
special provision being incorporated that the new stock when used as a 
re-recording negative must be capable of being printed onto present 
type release stocks developed in the standard manner. 

The interest in the use of fine-grain films for variable-density record- 
ing has been due to the possibility that these films might offer a con- 
siderable reduction in the background noise existing in present stand- 
ard stocks, and at the same time reduce the intermodulation of noise 
and signal. While considerable progress has been made in the use of 
noise-reduction circuits to enhance the apparent signal-to-noise ratio 
on standard stocks, it has been felt that a reduction in noise in the film 
itself is essential to accommodate fully the volume range obtainable 
in modern sound recording systems. Further, while the use of noise- 
reduction circuits has made possible an increase in apparent signal-to- 
noise ratio on standard stocks, the quality of the recorded sound 
tends to become impaired when noise reduction in excess of 10 db is 
applied, for beyond this point the breathing of the background noise 


tends to offset the benefits accruing from increased quietness during 
silent passages. 

Theoretical and experimental considerations have for some time 
led to the belief that reduction in the graininess of the negative and 
the positive sound stocks would assist in reducing background noise, 
and thereby enhance the signal-to-noise ratio attainable from sound- 
on-film; but until quite recently no fine-grain films that could be 
considered at all suitable for variable-density recording have been 
available. In 1937 a fine-grain sound emulsion was introduced by 
Eastman as type 1360 which, when exposed by white light, was in- 
tended for direct positive recordings as a substitute for the then cur- 
rent technic of ultraviolet exposure on standard sound negative emul- 
sions in the variable-area system. Dupont also introduced their 
type 216 for the same purpose. Both these films were primarily de- 
signed for variable-area work and with their inherently high contrast 
were not particularly suitable for variable-density low-gamma nega- 
tive development. Early in 1939 Dupont introduced type 222, re- 
ferred to previously, for variable-density negative and re-recording 
print use. The Eastman 1365, referred to previously, was actually 
introduced in 1937 as a fine-grain duplicating positive, but due to its 
low speed, was not considered at that time as suitable for variable- 
density recording. 

One of the problems confronting the use of any type of fine-grain 
emulsion for variable-density recording is the relatively low speed of 
these emulsions. In this connection it should be recalled that the 
use of standard sound emulsions in existing recording units calls in 
many cases for the maximum exposure obtainable from a tungsten 
lamp operated at its normal temperature. Consequently, with 
fine-grain films being offered with a speed of about one-tenth that of 
standard films, either a new light-source or drastic improvement in 
efficiency of old light-sources and optical systems becomes an ob- 
vious necessity. Fortunately, the development of the high-pressure 
mercury arc with a high intrinsic brilliance offers a possibility in 
this connection. The problem of securing sufficient exposure is not 
confined to the negative, but is also involved in securing sufficient ex- 
posure of the print. The high-pressure mercury lamp has been 
found to be the most suitable source for use in printing these stocks 
and several laboratories on the West Coast either are completely 
equipped or are in process of being equipped, with these light-sources. 

The high-pressure mercury arc as originally offered by the suppliers 


was not satisfactory as a sound recording medium due to the inherent 
high arc noise. After a series of cooperative tests between engineer- 
ing groups in the studios and laboratories in Hollywood, the lamp 
has been brought to a high state of efficiency and may now be con- 
sidered a satisfactory source of exposure for many of the existing fine- 
grain films. Further recent modifications have been made in this 
lamp by several of the Hollywood studios to provide greater illumina- 
tion than can be obtained by using it in its normal state. Thus, 
forced-air cooling of the arc (usually done by connecting a small intake 
tube at the end of the bulb and making a small opening in the outer 
envelope near the base and permitting a stream of air to pass the 
enclosed quartz arc) makes possible a considerable increase in il- 
lumination by operation at higher wattage than that for which the 
lamp was originally designed. This air-cooling device is being suc- 
cessfully employed at several studios and permits a range of exposure 
that could not have been obtained from the lamp without such air 

The use of the standard tungsten lamp as a source of exposure for 
these fine-grain stocks appears to offer possibilities in connection with 
the use of certain optical systems and under certain conditions of 
development. Thus, for some time M.-G.-M. have been using a 
standard tungsten source at or slightly above its normal temperature 
rating for all original recording, while the same studio employs the 
mercury arc for recording the release negative where a higher nega- 
tive density is required. Further, recent improvements in developers 
for use with low-gamma fine-grain negatives at the Paramount and 
M.-G.-M. West Coast Laboratories indicate that with improved re- 
cording optics, combined with ultraviolet printing, it may be quite 
feasible to substitute completely the use of a standard tungsten-fila- 
ment source for exposure of these films. 

Sufficient experimentation backed up by actual studio recordings 
have been made to indicate the order of magnitude of the improve- 
ments that may be expected from the adoption of fine-grain film, both 
as a variable-density negative and a positive. Thus, an improvement 
in signal-to-noise ratio, as measured in accordance with Section 6 
of the attached specifications, of at least 6 db is indicated on a fine- 
grain print made from a fine-grain negative. It should be pointed 
out, however, that the actual improvement to the ear seems more 
of the order of 8 to 10 db, this being accounted for largely by the al- 
most complete absence of breathing of the background noise. While 


one studio, using the 200-mil push-pull system, reports a satisfactory 
signal-to-noise ratio from fine-grain films without any applied noise- 
reduction, others using the standard system report that the use of a 
normal amount of noise-reduction may be associated with fine-grain 
film recording with highly satisfactory results. Laboratory measure- 
ments report that with a fine-grain stock, such as Dupont 222 or 
equivalent, a total signal-to-noise ratio of 45 db may be obtained. 
With the finer-grained Eastman 1365 this figure may be increased to 
about 48. Recording tests show that with the application of 10 db 
of noise-reduction to films of this nature, excellent quality of both 
dialog and music may be obtained. 

While the ultimate benefits from the use of fine-grain film can be 
obtained only by carrying the fine-grain all the way to the release 
print, considerable improvement in signal-to-noise ratio and in overall 
quality can be obtained by using it for original negative, re-recording 
print, and re-recorded negative, with the final print being made on a 
standard positive stock. In the latter case the noise and modulated 
noise effects appear to increase by 3 to 5 db in comparison with a fine- 
grain print. Considerable success has been found in the Hollywood 
studios from the use of fine-grain film for original negatives and re- 
recording prints only; the tendency, however, is to carry the fine- 
grain also to the re-recording negative stage. 

The improvements from the use of fine-grain film in variable- 
density recording are not confined to increased signal-to-noise ratio, 
but include other factors such as improved image definition which may 
be traced to reduced flare in the emulsion. This results in a clean- 
ness of all high-frequencies not hitherto attainable, and is also 
accompanied by a moderate increase in the high-frequency output 
in such films as compared to standard films. A very low degree of dis- 
tortion is indicated, when measured by intermodulation or harmonic 
analysis, on fine-grain prints made from fine-grain negatives, es- 
pecially if ultraviolet light is used in exposing the print stock. This 
low distortion undoubtedly is partially responsible for the pleasing 
quality of recordings made on fine-grain films. 

To obtain satisfactory results with current fine-grain stocks, dy- 
namic methods are recommended for determination of optimum proc- 
essing conditions, as misleading information may result from the ap- 
plication of classical sensitometry to these films. The use of these 
films under laboratory conditions generally calls for the exercise of 
greater care in processing and handling in order to avoid noise from 


dirt and abrasions. Such noise is more evident on these films due to 
lack of masking by the lower background noise inherent in these 
stocks. One of the chief problems presented with the use of these 
films has been that of being able to obtain a sufficiently low negative 
gamma, and at the same time obtain the required negative density. 
This has required a development of suitable negative baths. While 
the low speed of these films may still be considered a problem, the 
improvements in light-sources as well as in optics indicate that 
present fine-grain stocks or future ones, even more slow, may be ex- 
posed without any great difficulty. In spite of the difficulties at- 
tendant to the introduction of fine-grain film in the sound-recording 
field, the improvement in signal-to-noise ratio and in general quality 
mean then- inevitable introduction on a wide scale into the motion 
picture industry. 

JOHN G. FRAYNE, Chairman 
Fine-Grain Film Committee 

L. A. AICHOLTZ, Universal Studio 

F. G. ALBIN, Samuel Goldwyn Studio 

G. M. BEST, Warner Bros. Studio 

G. A. CHAMBERS, Eastman Kodak Co. 

C. R. DAILY, Paramount Studio 

J. K. MILLIARD, M.-G.-M. Studio 

W. W. LINDSAY, JR., G. S. S. I. 

H. W. MOYSE, Dupont Film Mfg. Corp. 

K. M. PIER, 20th Century-Fox Studio 

S. J. TWINING, Columbia 

ELMER RAGUSE, Hal Roach Studio 



These specifications cover a negative stock to be used for original recording 
and for release recording; also a print stock primarily intended for use as a re- 
recording print stock. It will be satisfactory if a single emulsion can satisfactorily 
fulfill the requirements as a negative and re-recording positive. 

These specifications do not apply to release positive film, although a number 
of the requirements listed below pertain to such a new film when it is designed. 

(1) Intended Usage. These films are intended for variable-density recording, 
assuming a white-light source of exposure for both negative and positive. 

(a) The negative is intended for original recording and re-recording operations. 

(b) The print stock is primarily intended for re-recording prints. 

(2) Gamma Infinity. (a) Negative stock gamma-infinity may be as low as 
1.0 and not greater than 2.0, assuming development in present variable-density 
sound-track negative developers. 


(fe) The re-recording print stock gamma-infinity should be such that a value of 
approximately 2.0 may be readily obtained with existing motion picture positive 

(5) Development Requirements. (a) The new negative stock when used for 
original recording must be capable of being printed onto the new re-recording 
print stock and attain a projected overall gamma of unity. 

(6) The new negative stock when used as a re-recording negative must be 
capable of being printed onto the present types of release stocks which will be 
developed in standard release positive developers; again attaining a projected 
overall gamma of unity. 

(c) The new original sound negative must be capable of being printed directly 
to the release positive stock with a projected overall unity. 

(4) Speed and Light Sources. For the purpose of this specification, the speed of 
variable-density sound negative will be defined as the log E value on an Eastman 
116 sensitometric scale, corresponding to the toe break of the H&D curve. 

(a) It is desirable that the speed of the negative be such that it may be cor- 
rectly exposed using unfiltered tungsten sources operating at color temperatures 
less than 3000K in connection with existing recording optical systems. If, how- 
ever, in order to meet the requirements of Section 6 below, it is found necessary 
further to reduce the speed, then a G. E. H3X mercury lamp, or equivalent, un- 
filtered may be considered as a source of exposure for the film. 

(6) Both unfiltered tungsten and unfiltered mercury lamp sources may be 
considered suitable for the printing operation. 

(5) Latitude. The latitude of the straight-line portion of the H&D curve 
used either as negative or positive should have a log E value of at least 1.3. 

(6) Signal-to- Noise Ratio. (a) Prints made on the new re-recording print 
stock from the new negative stock should show an increase in signal-to-noise 
ratio of 8 to 12 db over that now found in existing sound positives made from 
existing sound negatives. 

(6) Release prints made on the E. K. 1301 or Dupont 213 from the new sound 
negative should show an increase in the signal-to-noise ratio of 4 to 6 db over 
existing prints made from existing negatives. 

(c) Method of Measurement. The measurement of the signal-to-noise ratio 
for a given stock should be made, using a normal theater reproducing system 
whose frequency characteristic referred to constant modulation of the light in the 
reproducing aperture represents the ear-weighting characteristic equivalent to 
the 40 db loudness contour. The characteristic of this contour as adopted by the 
American Standards Association is as follows: 

F Relative Response 

60 -27.0db 

100 -19.0 

300 - 7.0 

500 - 3.5 

3000 + 3.0 

5000 +0.5 

8000 - 3.5 


The scanning width of the reproducing aperture is assumed to be the normal 84 

The maximum signal output of a print is defined as the reproduced 1000-cycle 
output, assuming 100 per cent modulation of the recording modulating device. 
The output at 100 per cent modulation may be obtained by extrapolation from 
the measured output obtained from a 50 per cent modulation of the recording 

The film-noise distribution throughout the spectrum should be relatively 
smooth, that is, it should not have a predominance of noise in any restricted fre- 
quency band. A check of such a condition can be made employing suitable band- 
pass filters. A qualitative check of noise may be made by projecting the films 
over a normal theater reproducing system. 

(d) Signal-to-noise ratios as measured above indicate values of 40 to 42 db for 
white-light prints on E. K. 1301 or Dupont 213 from E. K. 1553 or Dupont 214 
negatives developed in normal positive and sound negative developers. 

(7) Image Spread Halation. The film should be free of image spread when 
exposed under sound-recording conditions to either mercury arc or unfiltered 
tungsten light. In testing for image spread, the size and intensity of the light- 
source should simulate a light-valve as used for variable-density recording pro- 
ducing an image height of approximately 0.25 mil. Gray base or a fugitive 
dye may be used if necessary to obtain freedom from halation. 

(8) Image Stability. Over any period of time the new film should hold an 
exposed undeveloped latent image as well as E. K. 2559 or Dupont 214. 

(9) Durability. The wearing qualities of the film with normal handling should 
be at least as good as that obtained with E. K. 1369 or Dupont 214. 


MR. KELLOGG: Do I understand that the comparisons reported were between 
the ground-noise with the fine-grain film as compared with standard sound-record- 
ing stocks, when exposed with white light or ultraviolet? 

MR. MORGAN: They were exposed to unfiltered mercury arc light. 

MR. KELLOGG: If you have got to go to a mercury lamp to expose the fine- 
grain stocks, it might be reasonable to compare the fine-grain recording with 
standard film exposed with ultraviolet. The ultraviolet cuts down the graininess 
of standard recording film so much that there might be little to choose. 

MR. PALMER: So much progress has been made in improving the quality of 
sound by the use of these very fine-grain emulsions that it suggests the possibility 
that it might be advisable to record sound on a color-sensitive film in which the 
film image is a dyed image instead of a silver-grain image, and still further im- 
prove the quality of the sound. 

MR. MORGAN: That has been talked about. I do not know whether anyone 
has actually made tests. The thing that everyone is hoping that these develop- 
ments will accomplish is, of course, improved sound quality; but, in addition to 
that, a greater signal-to-noise ratio, making possible original recordings without 
the necessity of volume control. That would provide much greater latitude in 
the re-recording process; and even though it may be necessary to compress the 
volume range somewhat in the release print, a much better and more dramatic re- 


suit can be had if the originals are as near like the sound that occurred on the 
stage, in both volume and quality. 

MR. CRABTREE : I think this development indicates that if there is the remot- 
est possibility that any new idea will improve the quality of the sound or the 
picture, the Hollywood technicians are willing to expend the necessary time and 
money to try it out. This is very encouraging to those engaged in research who 
have proposed so many things in the past only to be literally sat upon because the 
new proposals involved changes in equipment or the expenditure of money. 

MR. MAURER: It appears that in the minds of most of us the principal reason 
for using fine-grain films is that they give reduced background noise. In this 
connection, and in view of what Mr. Crabtree has just said, I should like to call 
attention to one factor that has not been discussed here. That is the importance 
of dirt as a source of background noise when the noise due to film graininess has 
been largely eliminated. 

About nine years ago I was working on a problem that required the reduction of 
film background noise to an exceptionally low point. In the course of that work 
I made use of one of the earliest types of these ultra-fine-grain recording films. 
After having reduced the noise by all the ordinary technics known at the time, on 
standard film, I substituted the fine-grain film and found a reduction in noise of 
the order of eight decibels. 

Thereafter, somewhat by accident, I discovered that by taking altogether ex- 
ceptional precautions to obtain freedom from suspended matter in the water used 
to wash the films, and in the developing and fixing solutions, a further reduction 
in noise of the same order of magnitude could be obtained. But this improve- 
ment in background noise was not obtained by taking similar precautions in the 
processing of standard, or coarse-grain, films. 

I bring up the point at this time because the evidence from the work I did was 
so definite, so unmistakable, that, in spite of the fact that no one has reported any 
similar observations in the intervening time, I am still convinced that in a technic 
in which we reduce the noise that is inherent in the film itself to a low point, we 
ought to begin taking greater care than the industry has ever thought worth 
while in filtering the wash water and all chemical baths, and in the removal of 
dust from the air used to dry the films. 

DR. FRAYNE:* Answering Mr. Kellogg, measurements of signal-to-noise ratio 
on standard film exposed to ultraviolet light show an improvement of approxi- 
mately one db over that obtained in the same film when exposed to ordinary 
tungsten light. This is a negligible improvement compared to that reported 
from the use of fine-grain films. 

The points raised by Mr. Maurer have been given a great deal of attention by 
the Hollywood film laboratories and considerable success has been attained in 
reducing "pops" and other undesirable noises that were unmasked when the film 
background noise was reduced by the use of fine-grain film stocks. With regard 
to Mr. Maurer's claim that he successfully used an ultra-fine-grain recording film 
some nine years ago, it is a matter of regret that he did not proclaim this dis- 
covery from the housetops and thereby assist in solving one of the industry's 
most pressing problems. 

* Communicated. 


C. R. DAILY** 

Summary. Many types of picture scenes are improved in quality when some 
of the new fine- grain films are used as a printing stock. More detail on the screen 
and less image "boiling" is observed due to the greater resolution of the fine- grain films. 
When such films are used for variable-density sound recording, a material increase 
in volume range is obtained which permits greater latitude in the original and dubbing 
recording operations. The sound quality is improved due to the reduction in noise 
and modulated noise effects which partially mask the signal when the coarser-grained 
positive types of emulsions are used. Data are presented on some of the problems 
encountered in the use of fine-grain films for dubbing prints, release negative, and 
release prints. 

The need of finer-grained film stocks for release use has been felt 
for some time because of the recognized limitations of the older posi- 
tive type emulsions. This paper describes the methods recently 
developed and perfected at Paramount Pictures, Inc., which make 
possible the commercial use of one of the new fine-grain emulsions 
for the release sound negative and for the movietone release picture 
and sound print. The release use of fine-grain films provides a picture 
image which is improved in definition, the signal-to-noise ratio on the 
sound print is materially increased and the dramatic value of recorded 
material enhanced by the reduction of modulated film noise. The 
completion of this important development now makes possible the 
full utilization of the benefits of fine-grain films for original and re- 
lease sound-track and release picture prints, in addition to the present 
use of such films in making improved master positives, duped nega- 
tives, and projection background prints. 

Representatives of the sound departments of a number of the 
studios were instrumental in forming a committee in January, 1939, 
to draw up specifications for fine-grain films better adapted to variable- 
density recording than the fine-grain films then available. This 

* Presented at the 1939 Fall Meeting at New York, N. Y.; received October 
14, 1939. 

** Paramount Pictures. Inc., Hollywood, Calif. 



Committee, comprised of representatives of E. R. P. I., Dupont, 
Eastman, M.-G.-M., Paramount, Samuel Goldwyn Studio, Universal, 
Twentieth Century-Fox, G. S. S. I., Columbia, and Warners, pre- 
pared over a period of time a set of specifications which were sub- 
mitted to the film suppliers. Since a progress report of the activities 
of this Committee is being presented in a separate paper, 1 it will not 
be covered here. While most of these organizations initially re- 
stricted their experimental work to the application of fine-grain 
films for original recordings and dubbing prints, the activities at 
Paramount were independently directed toward the specific problem 
of applying such films for use as a release sound negative and a 
release picture and sound print, since it was felt by this studio that 
only by such an application to release printing could the theater 
patron be benefited to the greatest possible extent, in regard to both 
picture and sound. 

Emulsion Limitations. The positive type of film emulsions cur- 
rently used for variable-density sound negatives, dubbing prints, 
and release prints have a relatively coarse-grain structure. As a 
result, sound prints are subject to a considerable amount of film 
noise, modulated film noise effects, loss of high-frequency resolution, 
some image spread and halation, and other problems which impair 
the realism of sound recordings. Picture prints are likewise restricted 
in definition and clarity. On account of the excellent results obtained 
with fine-grain films for duplicating and projection background 
prints, it was logical to expect that a similar improvement would be 
obtained if suitable films of this type were applied to variable-density 
sound recording and release picture printing. 

Minimizing Film Noise Effects. Since the basic film noise of posi- 
tive type emulsions has remained essentially unchanged since the 
advent of sound pictures, the equipment suppliers and studios have 
developed a number of devices to reduce the disturbing effects of this 
noise, such devices including noise-reduction equipments, push-pull 
recordings, track-matting devices, wider original sound-tracks, and 
pre- and post-equalized original push-pull recordings. Ultraviolet 
printing of variable-density sound-track on positive type emulsions, 
has also been found to reduce slightly the apparent film noise, at the 
same time improving the high-frequency response and reducing the 
halation effects. Of the film noise reduction methods pointed out 
above, ultraviolet printing is the only one which basically affects the 
source of film noise and this to only a limited degree. 

14 C. R. DAILY tf. s. M. P. E. 

Initial Applications. Fine-grain duplicating type films were first 
made available in 1937. They were immediately applied to make 
master positives, duped negatives, and projection background prints, 
with a material improvement in quality. The first experimental work 
on fine-grain variable- density dubbing prints was also undertaken at 
this time. During the past two years, fine-grain films have also been 
applied to a limited extent as an original recording and dubbing print 
stock for variable-area. Since the above-mentioned uses of fine- 
grain films have been adequately covered in the literature, the dis- 
cussion in this paper will be limited to the recent application of fine- 
grain films to variable-density release recording and movietone sound 
and picture release prints. 

General Requirements for Release. The following general require- 
ments were indicated as desirable in the fine-grain films which would 
prove of maximum use to this studio at the present time: 

(1) Normal release positive developer, with development at ap- 
proximately the normal time, to be used for the proposed fine-grain 
movietone release and for fine-grain dubbing prints. 

(2) Higher-intensity-printing light-sources to be made available 
if necessary. 

(5) Usable as a release sound negative, requiring special exposure 
and development if necessary. 

(4) These requirements indicated a stock which would have a 
suitably high gamma in the positive developer for the release picture 
and, in addition, could be developed to a suitable low gamma for the 
release sound negative. 

In addition to the above requirements for the handling of such 
films, the following benefits were desired: 

(1) Picture detail to be improved and the projected color to be 

(2) Release sound-track print noise to be materially lower than 
that obtained with the currently used combinations of positive type 

(5) Frequency characteristic equal to or superior to that obtained 
with positive type emulsions. 

(4) Overall dynamic distortion of the sound-print to be reduced. 

Tests at a number of studios during the early part of 1939 on ex- 
perimental and existent fine-grain emulsions led to the development 
of the Dupont 222 type emulsion. Since tests on this particular stock 

Jan., 1940] 



indicated that, in general, it fulfilled the requirements listed above, 
experimental work at their studio was concentrated on that particular 
emulsion in order to work out all the phases of operation necessary 
for its early application to release. The slowness of the emulsion 
and apparently greater susceptibility to abrasion and to picking up 
dirt, however, are problems involved in its use. As would be expected, 
it is necessary to exercise greater care in keeping the negative clean 
since the reduced general film background noise makes printed- 
through dirt noise much more apparent. 

/ o 




Loo f 

IG. 1. 116 sensitometric characteristics of Duppnt 222 
fine-grain and Ek-1301 positive emulsions in release positive de- 
veloper; same development time. 

The following paragraphs cover some of the technical aspects of the 
present use of this particular emulsion as applied to release. 

Printing. As an illustration of the lower printing exposure speed 
of fine-grain film, standard lib sensitometric characteristics are 
shown in Fig. 1 for the 222 and 1301 emulsions when developed at 
the same time in release positive developer. The circles indicate the 
approximate visual print densities used for the unbiased printed- 
through sound-track, (a) for the case where 222 is used as a negative 
and as a print, and (6) for the case where 1359 is used as a negative 



[J. S. M. P. E. 

for 1301 prints. The sensitometric A log E of 0.9 indicated to obtain 
these densities, corresponds to a ratio of eight in exposure speed. 
In order to obtain this increase in printing light, Bell and Howell 
Model D printers were first modified by replacing the 165- watt 
tungsten lamps with General Electric H3X mercury arcs operating at 
approximately 90 watts. Adjustments of the wattage and ground- 
glasses provided the necessary control to obtain the desired mean 
exposure. These arcs have proved to be stable in operation and have 

a long life characteristic. While Model 
E release printers require a still higher 
intensity light-source, it now appears 
practicable to equip these machines with 
suitable mercury arcs for fine-grain 

Release Recording. Fine-grain films 
for release sound negative use also re- 
quired an increase in the intrinsic bril- 
liance of the recorder lamp. In addition, 
the Western Electric D-86715 film re- 
cording machines used at this studio for 
release recording are equipped with 
track-matting devices which still further 
reduce the exposure obtainable, by com- 
parison with production recorders, on ac- 
count of the additional optical system 
employed. Tests on the coiled-filament 
tungsten lamps normally employed in- 
dicated that insufficient exposure could 
be obtained at a safe operating current. 
The General Electric H3X 85-watt high- 
pressure mercury arc was therefore tested as a new light-source and 
was found to have the following characteristics: 

(a) At 90 to 100 watts, the exposure obtained was just sufficient 
with no margin of added exposure which might be needed to take care 
of variations in stock and developer sensitivity. 
(6) Warm-up time of 5 to 10 minutes required. 
(c) Cooling time of approximately 5 minutes required if the arc 
were accidentally extinguished, before it could be restruck, with a 
further delay due to the new re-heat time required. Such delays 
might be quite expensive to production and to overcome them might 

FIG. 2. (a) H3X arc 
modified for forced-air 
cooling; (ft) Normal H3X 

Jan., 1940] 



require the operation of a spare lamp in a pre-focused mounting. 
Development work on the H3X arc, modified for forced-draft air- 
cooling, indicated a number of advantages: (a) Twice the normal 
exposure obtainable, (b) starting time reduced to one minute or less, 
(c) no delay in restriking the arc if accidently extinguished, (d) arc 
self-extinguishing if air or power supply interrupted. Views of the 
normal and modified arcs are shown in Fig. 2. A glass tube is sealed 

FIG. 3. Exposure characteristics of light- 
sources for release type Western Electric 
Recorder. Dupont 222, special negative de- 
veloper, gamma 0.28; (a) air-cooled H3 X 
arc; (b) H3X arc; and (c) 9.5-v, 9-ampere 
coiled-filament tungsten lamp. 

to the end of the envelope to permit connection to the compressed- 
air line, and a small hole added in the envelope near the base of the 
tube to permit escape of the air. 

Fig. 3 shows a comparison of the relative exposures obtained on the 
release recorder using (a) the normal 9.5-volt, 9-ampere coiled-fila- 
ment tungsten lamp, (b) the unmodified H3X arc, and (c) the air- 
cooled H3X arc. Dupont 222 stock was used for this test, developed 
to a control gamma of 0.28 in the special borax negative sound de- 
veloper which had been made up for this work. A visual negative 



[J. S. M. P. E. 

density of 0.38 to 0.40 was indicated as optimum from distortion data 
which will be explained later. In order to obtain this density, it would 
be necessary to operate the 9-ampere coiled-filament lamp at 10.4 
amperes, a value too high for safe commercial use. The normal H3X 
arc would be operated at 80 to 100 watts with little margin of safety 
since these lamps swell and become unusable at higher wattage. 
The air-cooled arc operating at 120 to 140 watts provides adequate 
exposure with a margin of safety of at least two times. Tests indicate 
that it can be operated for considerable periods of time at 200 to 300 
watts with only a gradual deterioration of the cathodes. 



RetArtrf Exposure C/MXAcreittaTtc 



Alft C<MM.0 H3X AtK 













H o 

IOC 200 JOO 4OC 

D/aaiPAreo /N flip COOLMO H3X 
FIG. 4. Exposure of air-cooled H3X arc referred to a normal H3X 
arc operated at 85 watts. 

The relative wattage-exposure efficiency of the air-cooled arc vs. 
the normal arc is shown in Fig. 4, the data being derived from Fig. 3. 
This curve indicates that the air-cooled arc required approximately 
50 per cent more wattage to be dissipated in order to produce the 
same exposure as the normal arc operated at 85 watts. The simpli- 
fied circuit of the air-cooled arc supply is shown in Fig. 5. A 200- 
volt generator is connected to the arc through a filter circuit and ad- 
justable series resistance R. The series inductance termination toward 
the arc was found desirable to eliminate an occasional tendency of 
the arc to oscillate at very low frequency when very little series re- 
sistance was included in the circuit, and the filter section was termi- 

Jan., 1940] 



nated by a capacity. The required starting voltage for the arc is 
being temporarily supplied from a high-voltage rectifier which was 
available. This starting circuit can be replaced by a high- voltage 
transformer or other means for striking the arc. The switch Su>i is 
normally closed while the arc is in operation. Since the release re- 
corders are permanently installed, the lot supply of compressed air 
has been used. A suitable filter inserted in the air-line prevents con- 
tamination of the surface of the arc tube and the inside of the pro- 
tecting glass envelope. An Airco regulator and pressure-meter con- 
trol the air-pressure. It is important that the flow of air passing the 

Conrtrjaro At* iittf ^-^ 

FIG. 5. Control circuit of air-cooled H3X arc. 

air-cooled arc be quite stable; otherwise erratic fluctuations in the 
light output will occur. 

The method of starting and operating the air-cooled arc as de- 
veloped here appears to offer definite advantages over other modes of 
operation which were tested. The voltage across the arc is not per- 
mitted to exceed 150 volts and it is normally operated at 125 volts. 
This restriction, together with precision control of the air supply, 
has resulted in a long life with stable exposure. 

To start the arc, the air supply is turned off, the resistance R set 
at a fairly high value, and the switch Swi momentarily opened. 
The arc starts operating at approximately 0.5 ampere at 10 to 20 
volts. R is then reduced until the current approaches 1.5 amperes, 
the voltage across the arc being approximately 20 volts because the 



Lf. S. M. P. E. 

arc is still fairly cold. During the next 20 to 30 seconds the voltage 
will rapidly rise, R being adjusted to keep the current less than 1.5 
amperes. The air is turned on across the arc when the voltage 
equals 100 volts, the voltage across the arc never exceeding 150 volts. 
Simultaneous adjustments are then made of the resistance and the 
air supply so that the voltage is approximately 125 volts and the 
wattage approaches the desired value as indicated on the wattmeter 
W. One arc which has been in service for some time has been started 
in this manner more than 100 times with no appreciable deterioration. 
It is normally operated at approximately 140 watts, 125 volts, 1.1 
amperes. The air-cooled arc is mounted on the recorder in a special 
bracket and is enclosed in a protective housing to prevent erratic 

FIG. 6. Dupont 222. Time-gamma characteristics in special 
negative developer. 

cooling from the outside air sources. An arc-welding type glass 
filter is mounted in the housing to permit ready inspection of the 
lamp without damage to the eyes of the operator. 

New Negative Developer. Since the objective of this development 
was to provide a satisfactory fine-grain release positive and a fine- 
grain release sound negative, "the sound negative would have to be 
developed to a low gamma if a high-quality sound print were to be 
obtained. Tests on Dupont 222 in the normal sound negative de- 
veloper indicated a control gamma of approximately 0.5 at the short- 
est time of development. Since the distortion obtained at this gamma 
was too high when printed with a mercury arc to Dupont 222 release 
stock, the group at the Paramount Film Laboratory, under the direc- 
tion of Messrs. J. R. Wilkinson and F. L. Eich, engineered a special 
borax sound negative developer which has proved to be very satis- 
factory for the proper low-gamma development of Dupont 222. 

Jan., 1940] 



Fig. 6 shows the II& sensitometric time-gamma characteristics of 
222 stock in this new developer, the developer being used in a normal 
sound negative developing machine. A visual control gamma of 
0.27 has been found to be optimum for the 222 negative when prints, 
as indicated above, are made on 222 stock and handled in the normal 
release developer. 

Fig. 7 is shown to indicate the difference in density of sensitometric 
exposures obtained on (a) Dupont 222 when handled in the special 
developer and (6) EK-1359 sound negative stock developed in the 
normal sound negative developer. The operating visual control 
gammas are 0.27 and 0.35, respectively. An increase in sensito- 

EK-/3S9 i, DufoMr ZJUt 

FIG. 7. 116 sensitometric characteristics of (a) EK-1359 in 
normal negative sound developer; (&) Dupont 222 fine-grain in 
special negative sound developer. 

metric exposure of approximately seven times is indicated in order to 
obtain the desired operating density on the fine-grain stock. The 
circled operating points are the approximate normal values used in 
production. The 1359 negative is normally printed with white- 
light to 1301 while the 222 stock is printed with a mercury arc to 222, 
both prints being handled in the normal positive developer. 

Processing Controls. Tests on 222 as a picture print stock indi- 
cated that improved quality prints could be obtained by developing 
this stock at the normal time in the release developer. Therefore, 
with the processing characteristics of the print established, it was 
necessary to find the optimum constants for the negative, and the 
optimum print density. To facilitate this determination, use was 
made of the intermodulation test equipment recently developed by 
Electrical Research Products, Inc. 2 With the aid of this equipment 



[J. S. M. P. E. 

it is possible to determine readily the dynamic distortion of sound 
prints, using standard theater type reproducing equipment with the 
measuring apparatus. This tool is of considerable advantage in 
variable-density film research since no corrections must be applied. 
Static sensitometry provides the necessary laboratory controls but 
since the source of light and methods of exposure are not the same as 
in a recording machine, it does not give the true projection char- 
acteristics of the film. 

In order to determine the processing constants of the fine-grain 
negative which would be best adapted for fine-grain printing, static 

FIG. 8. Intel-modulation distortion curves for (a) fine-grain 
release prints; (b) fine-grain dubbing prints; and (c) normal 
positive stock release prints. 

sensitometric tests were first made of the new developer. 60-400 
cycle intermodulation tests were then recorded at approximately 80 
per cent modulation of the light- valve. This test was then repeated 
several times to provide a family of negatives covering the range of 
negative densities and gammas to be investigated. Prints of these 
families of negatives were made at various densities and measured 
on the distortion-analysis equipment, the results plotted, and the 
optimum operating points determined. This test serves inde- 
pendently to define a satisfactory negative exposure, negative control 
gamma, and positive print density, the print gamma being fixed by 
the positive developer. 

Jan., 1940] USE OF FlNE-GRAIN FlLM 23 

Tests made in the same manner also disclosed that dubbing prints of 
materially improved quality and lower surface noise could be made 
on 222 stock using the normal low gamma 1359 negative. 

In Fig. 8 the average values of intermodulation distortion as a 
function of visual unbiased print density are shown for the combina- 
tions of stocks currently used. The processing constants for the nega- 
tive are optimum in each case. 

(1) Fine-Grain Release. Dupont 222 negative, air-cooled H3X arc recorder 
exposure, visual negative density 0.38, control gamma 0.27, special developer; 
Dupont 222 Hg arc print, normal positive developer. This improved combina- 
tion is also applicable to original recording, and if made in this manner the original 
sound negative could be inter-cut with dubbed release negative. 

(2) Fine-Grain Dubbing Prints. EK-1359 negative, tungsten lamp expo- 
sure, visual negative density 0.55, control gamma 0.35, normal sound negative 
developer. Dupont 222 Hg arc print, normal positive developer. 

(5) Positive Type Emulsion Negative and Prints. This combination repre- 
sents the former normal operation with the coarser-grained film stocks. EK-1359 
negative handled in the same manner as in (2); EK-1301 tungsten print, normal 
positive developer. 

The optimum values of intermodulation distortion of 7 to 8 1 /* 
per cent indicated in Fig. 8 correspond to single low-frequency total 
harmonic distortions of approximately 1.8 to 2.2 per cent. 2 

Noise. The signal-to-noise improvement was quite marked when 
fine-grain stocks were used. Table I shows the order of magnitude 
of the reduction in film noise which has been obtained. 


Signal-to- Noise Ratios Referred to Positive Type Emulsions 

Negative D-222 EK-1359 EK-1359 

Print D-222 D-222 EK-1301 

Use Release or Dubbing Release or 

Original Prints Original 

Relative Noise 6 to -8 db -3 to -4 db db Reference 

These determinations were made by aural comparison of unmodu- 
lated sections of the various types of tracks. However, it should be 
pointed out that for many types of recorded material, the reduction 
of modulated noise effects appears to be still greater. The lower 
noise level removes a veil which formerly masked many sounds, in- 
creasing the clarity and definition of the recorded signal. As these 

24 C. R. DAILY [J. S. M. p. E. 

improved stocks come into more general use, greater precautions will 
have to be taken to reduce extraneous stage noises, particularly dur- 
ing quiet, intimate scenes. 

The greater volume range available increases the mixing latitude 
available to the original and re-recording mixers, thereby reducing 
the variations in background noise in the release product. At the 
same time a reduction in recording levels may be possible to re- 
duce the number of recorded overloaded peaks. Experience has 
indicated that only a small fraction of all speech peaks are in the top 
6 db of the volume range ; therefore, since the present recording level 
has been set by limitations of film background noise on the one hand, 
and modulator overload on the other, the reduction of film back- 
ground noise will permit the recording at a lower level of a cleaner 
track with no increase in noise in the final product. 

The use of fine-grain films as negative or as print also effects a small 
improvement in the high-frequency response. The probable future 
use of ultraviolet printing of fine-grain stocks still further improves 
the high-free uency response, at the same time minimizing halation 
effects which are still present to some extent with the present technic 
of printing with unfiltered mercury arc light. 

Conclusions. The commercial application of fine-grain film stocks 
for release sound and picture printing, for release sound negative, 
and for dubbing prints has effected material improvement in picture 
detail and sound quality. The volume range has been considerably 
increased and the disturbing effects of modulated film noise 


1 FRAYNE, J. G. : "Report on the Adaptation of Fine-Grain Films to Variable- 
Density Sound Technics," this issue of the JOURNAL, p. 3. 

* FRAYNE, J. G., AND SCOVILLE, R. R.: "Analysis and Measurement of Dis- 
tortion in Variable-Density Records," /. Soc. Mot. Pict. Eng., XXXII (June, 1939), 
p. 648. 


MR. FRIEDL: There was definitely a yellow tone in the picture shown on fine- 
grain stock. Of course, that is nothing new. We are all acquainted with those 
characteristics of fine-grain films, and we want to make sure in improving the 
sound that we do not work to the detriment of the white-colored screen image 
that everybody is working for. We must not fall back to low-intensity lamp days 
in trying to achieve perfection in sound. Motion pictures are a combination of 
picture and sound and we want to develop both together. 

Jan., 1940] USE OF FINE-GRAIN FILM 25 

MR. GRIFFIN: The difference in color in these two presentations of this same 
scene were so marked that I had hoped there would be quite some discussion with 
regard to it. 

MR. CRABTREE : It would be interesting to know which they prefer. Do they 
prefer the blue-black or the warm tone? (Ed.: A show of hands indicated 
approximately a 50-50 preference for the two tones.) 

MR. SHULTZ : The projection equipment used here might be low-intensity. 

MR. GRIFFIN : It so happens that it is. 

MR. SHULTZ: Would not that tend to produce a yellowish appearance? 

MR. GRIFFIN: We did notice, notwithstanding that, that one of the prints was 
distinctly yellow. Due to the small size of the picture, notwithstanding the type 
of arc used here, we are getting high illumination on this screen. If the screen 
were much larger, the difference would be more pronounced. 

MR. KELLOGG : There is a chance for a little confusion in comparing the brown- 
ish effect that may result from the print with that related to the type of light- 
source. One tends to alter the color of the highlights, and the other affects the 
tones of the shadows. 

MR. GRIFFIN: That is true, but it must be remembered that both prints were 
projected with the same light-source, and there was considerable difference in 

MR. DAILY:* One complete feature-length picture has already been printed 
on this type of fine-grain stock and has been reviewed in a number of theaters 
and review rooms having high- and low-intensity arcs. The soft ivory tint 
noted in the picture has caused favorable comment because of reduced eye- 
strain and the increase in resolution which is obtained. The reduced image 
boiling on the screen makes the front seats in the theater much more usable 
from an audience standpoint. As mentioned by Mr. Kellogg, it is true that the 
color characteristics of all films are a combination of the characteristics of the 
type of film stock used and the type of projection light-source. Fine-grain films, 
as well as normal positives, benefit by the use of high-intensity arcs, and it is 
hoped that their use will rapidly become more general. 

* Communicated. 


Summary. Release print laboratories require a method of quickly making 
photographic duplicate negatives to replace damaged original negative sections. 

The paper shows the development of a method of making dupes which have the same 
optimal print densities as the original negatives. A selection of film stocks is first 
made using the ratio of transmission between black area and fogs as a "yardstick." 
Processing data are shown in a family of curves, and the "best operating point" for 
the entire process has been determined. Fog considerations have been carried through- 
out the process. 

An outline of commercial practice is given together with the necessary routine 

In release print laboratories, it is necessary to have some method 
of quickly making duplicate sound negatives to replace damaged 
original negative sections. At present, re-recording and photo- 
graphic duping are the two methods of making duplicates in use. 
In the re-recording process, a carefully preserved release print or 
master sound print from the original negative is run in a film phono- 
graph and the sound re-recorded with a standard variable-area or 
variable-density recording machine. However, the making of 
duplicate negatives, or dupes, by this method is not practical unless 
the laboratory is connected with a studio where recording apparatus 
and trained personnel are available. It would be far too costly 
for a laboratory to maintain recording equipment for the sole purpose 
of making duplicate negatives. 

Photographic dupes have probably been in use as long as re- 
recording dupes, but the technic in general use and the resulting 
sound quality did not exactly suit our needs. It was decided, there- 
fore, to develop a process of making variable-area dupes using the 
cross-modulation test method of determining best processing condi- 
tions. The following criteria were set for the process. 

The quality of the sound from the dupe negative had to be high, 

* Presented at the 1939 Fall Meeting at New York, N. Y. ; received September 
19, 1939. 

** Ace Film Laboratories, Inc., Brooklyn, N. Y. 



so that a trained observer, when listening to a print, would have 
difficulty in telling which section had been printed from a dupe. 

All developing had to be done in the regular release print positive 
bath, at standard developing time. This requirement was set to 
save time, for inasmuch as the regular positive bath is in constant 
use no special machinery need be started to develop a dupe. Then, 
if dupes develop for the same time as standard prints, they may be 
spliced to the end of a release reel with a further saving of time and 
with no disruption of the regular work. 

The dupe negative had to have the same optimal print density 
as the original negative, and the same fog in the clear areas, in order 
that the inserted dupe negative might be printed on the same printer 
light as the original negative. In other words, the dupe should call 
for the same printer light as the original. This criterion was set 
because the printing machines in use had no method of changing 
the printing light within a reel. 

From the above it is seen that a "one to one" process was required. 
That is, we wanted a method of making a new negative which would 
have all the characteristics of the original negative from which it 
was to be made; it had to have the same fog and degree of image 
spread as the original negative, hence the term "one to one." 

In general, the dupe process was to follow standard procedure, in 
that a master or dupe positive was to be struck off from the negative 
before any release prints had been made. This was to be stored for 
use in the making of dupe negatives when the need arose. The print- 
ing of the master sound from the original negative, and the printing 
of the dupe negative from the master sound print was to be done on 
a non-slip contact printer. 

In order to satisfy the criterion of low fog or low density in the clear 
portion of variable-area dupe negative sound-track and yet have 
the dark area sufficiently black, it is necessary to maintain a fairly 
high contrast in every step of the process. However, the concept of 
high contrast or high sensitometric gamma is not enough. Gamma 
is the slope of a density-log exposure curve which in itself tells little 
of the ratio of transmission between the blacks and the clear areas 
on variable-area sound; and in printing it is this transmission ratio 
that is important. Prints are made through the clear areas and the 
blacks act as masks. 

An approach to this angle of the problem was made through the 
transmission ratio chart of Fig. 1. These curves, each of which 



[J. S. M. P. E. 

represents a density of variable-area sound, were calculated from the 
following formula: 

Transmission Black Log" 1 Density Clear 
Transmission Clear Log" 1 Density Black 

This family of curves shows the necessity for low fogs and high 
densities. High densities require a film stock with a high gamma, 

J>* At .06 .10 -/2 J* .it J4 .to .21 . 
OCNi/TY tf Of** KM A 

FIG. 1. Transmission ratio chart. 

and low fogs result from a combination of high gamma and short 
toe characteristics. Therefore, in order to study these two factors, 
sensitometric curves of several film stocks were made, three of which 
are shown in Fig. 2. The stocks shown are all of Eastman manufac- 
ture and are respectively 1363 high-contrast positive, 1301 regular 
release print positive, 1365 fine-grain duplicating positive. The 
tests shown were exposed and developed as follows: 

Jan., 1940] 



1363 and 1301 exposure with ultraviolet light on lib Eastman Sensitometer, 
developed 3Vi minutes in a positive bath. 

1365 exposure with white light on lib Eastman Sensitometer, de- 

veloped 7 l /z minutes in a borax negative bath. 

This particular developing time for 1365 stock is that which gives 
best picture quality at our laboratory, and the curve was made in 
order to determine whether master positive sounds could be satis- 

FIG. 2. Sensitometric curves of film stocks. 

factorily printed on the same film with the master positive picture. 
The curve shows that it is rather difficult to secure high densities on 
this stock and that the toe is rather long. These two disadvantages, 
coupled with the fact that present non-slip contact printers tend to 
scratch the picture area of a negative, made it advisable to use a sepa- 
rate film with a high contrast-ratio for the master positive sound. 

To test these two stocks, 1301 and 1363, the transmission ratio 
curves of Fig. 3 were made. Short prints, each a few feet in length, 
were made on an ultraviolet non-slip printer at several densities from 



[J. S. M. P. E. 


an original negative whose density was 2.12 and fog 0.05. It will 
be seen that the transmission ratio rises much faster than the fog. 
In the past, it has been somewhat customary to hold the bjack den- 
sities down to avoid excessive fog in a dupe positive, but it is now seen 
that the high densities give high transmission ratios in spite of rising 
fogs; and high fogs are no disadvantage in a master positive. A 
fog merely means that more light must be used to print the dupe 
negative. There seems to be little choice between the two films. 

It was decided to use 1363 as 
a master or dupe positive because 
it was felt that its high contrast 
would give a sharp boundary 
between clear and dark areas of 
track ; and it was further decided 
to attempt to use 1301 as a dupe 
negative. Fig. 4 shows a cross- 
modulation survey of a duping 
process using these two stocks. 
To make this family of curves, 
the procedure was as follows. 
Five master positive prints, each 
about 15 feet in length, were 
made from an original nega- 





FIG. 3. Transmission ratio chart 
of 1301 and 1363 stocks as a function 
of fog. 

tive cross-modulation test (9000 
cycles modulated with 400 cycles). 
A range of densities from ap- 
proximately 2.0 to 2.5 was 
picked after a study of Fig. 3 showed that the transmission ratio 
rose even with increasing fog. From these five dupe positive cross- 
modulation tests, 19 dupe negatives were printed ranging from ap- 
proximately 1.40 to 2.35. Seven prints were then made from each of 
the 19 dupe negative cross-modulation tests, in order to determine 
the cross-modulation print optimum of each dupe negative. Seven 
points are about the minimum number which will give a clearly 
defined cross-modulation optimum. This step resulted in 19 times 
7, or 133 prints, each 15 feet long. These were reproduced through 
a 400-cycle band-pass filter and the point of minimum 400-cyde 
output determined for each of the 19 print groups. 

Fig. 5 shows two typical cross-modulation curves, one represent- 
ing a print from an original negative, and the other, a print from a 

Jan., 1940] 



dupe negative. The ordinates are the number of decibels below a 
reference frequency of 400 cycles, the amplitude of which was the 
same as that of the 400-cycle frequency which modulated the 9000- 
cycle note in the recording of the original cross-modulation negative. 
The optimal points are those print densities where the curves have 
minima. Twenty such curves were made including the original, 
but only two are shown because of lack of space. 


1.10 ISO 1.70 1.80 l.$0 2.00 2.10 2.20 2.30 

FIG. 4. Family of curves showing processing optima 
for a duping process using 1363 as a master positive and 
1301 as a duplicate negative. 

Fig. 4 shows the result of all the cross-modulation tests plotted in 
the form of a family of curves, a family being necessary because we 
wanted to represent three variables, namely: density of the master 
or dupe positive, density of the duplicate negative, and optimal print 
density. Each curve shows a group of negative densities and their 
resulting optimal print densities all of which were made from one 
dupe positive density. There are five such curves representing the 
original five master dupe densities. 

One of the original criteria was the aim to have the dupe negative 



[J. S. M. P. E. 

optimal print density the same as that of the original negative whose 
optimum of 1.63 is shown in Fig. 5. This value of 1.63 is shown in 
Fig. 4 as a dotted line, and the above consideration of identical 
optima is satisfied for any points of intersection with the family of 
curves. We, thereby, find five combinations of dupe negative and 


11,0 iyo 1.60 1.70 i.eo 1-90 


FIG. 5. Two typical cross-modulation print curves 
showing optima, and 30-db cancellation density range. 

master positive densities which yield print optima identical with the 
original. These points are plotted in Fig. 6 which shows, for in- 
stance, that when the dupe positive is printed to a density of 2.20, 
the dupe negative should be 2.06. 

Any combination on the curve of Fig. 6 will fulfill the criterion of 
identical dupe print and original print optima, but it was felt that 
one particular combination, out of all those shown in Fig. 6, probably 

Jan., 1940] 



gave the best results ; or, perhaps our family of curves was not suf- 
ficiently extensive to have hit the best possible operating conditions. 
So, in order to determine the best possible combination of master 

2. IS 




2.10 2.20 ZJO 



FIG. 6. Combinations of dupe positive and dupe 
negative which give the same print optima as the original 


ti w 


<* 10 



pyp/r NfeATiVE 



Z.IO 2.20 2.30 


2. *0 


FIG. 7. Cross-modulation level as a function of combina- 
tions of dupe negative and dupe positive densities. 

positive and dupe negative densities, it was decided to use as a factor 
of merit the number of decibels which the dupe print optima were 
below the reference level. Fig. 5 shows that this figure for one par- 
ticular optimum is 43 l /z db. Fig. 5 also indicates the 30-db can- 



tf. S. M. P. E. 

cellation points which are the print densities at which the 400-cycle 
cross-modulation output rises to the level of 30-db below the refer- 
ence frequency. It is generally accepted that any print whose 
cross-modulation level lies below 30 db is satisfactory, so the width 


1.80 2.00 



FIG. 8. Fog curve of dupe negative as a function of 
dupe positive and dupe negative density. 

of the curve in the parameter of density measured between the 
intersection points with the 30-db line is an operating range for the 
print. The greater the width the greater the print density tolerance, 
so this width which we have called "density range" is also a factor 
of merit for the process. 

FIG. 9. Fog in final print as a function of combinations 
of dupe positive and dupe negative densities. 

The factors of merit of "db down" and "density range" are known 
for all the 19 points of Fig. 4. It is an easy matter to interpolate 
between these points and find the values of the factors of merit for 
every point of Fig. 6. These two factors of merit are shown in Fig. 
7 where the curve of Fig. 6 has been swung down to the X axis to 
form the double abscissa shown. It is very gratifying to find that 

Jan., 1940] 



there is a "best operating point," and that at this point, the cross- 
modulation level is down 43 l / 2 db, or l l / 2 db poorer than the cross- 
modulation level of the original print. The "density range" at this 
"best operating point" is 0.23 which compares very favorably with 
the density range of the original of 0.28. 

Fig. 8 shows a family of curves which represent the fog or clear 
area densities of the dupe negative for all the points shown in Fig. 4. 


& ^ C; * C; 

S Ci o o 3 




4L Ot 

























I.7O 1.80 l.<)0 2.00 



FIG. 10. Curve of print optima vs. dupe negative 
density. All negatives printed from a dupe positive 
density of 2.05. 

Throughout this paper all densities are given excluding densities of 
base, unless otherwise specified. The point represented by the large 
circle is the "best operating point" determined from Fig. 7, and the 
fog of 0.06 at this point is very satisfactory when compared with 
the fog of 0.05 of the original negative. 

Fig. 9 is a curve having the same abscissa as Fig. 7, but its or- 
dinates are clear side densities or fogs in the final print. These fogs 
are low enough to be negligible over the entire range. 


All these data, so far recorded, represent a treatment of one par- 
ticular cross-modulation test negative taken from one reel of a feature 
picture. Now, it is obvious that it would be impossible to treat every 
reel of every feature in the above manner, but such complete treat- 
ment is not necessary, for the procedure developed above is a "one 
to one" process. That is, for all practical purposes, it duplicates the 
image-spread of the original negative. However, at no place in the 
process does the degree of image-spread of the original negative enter; 
so we feel that within reasonable limits the above-determined "best 
operating point" holds for any original negative, providing the printer 
characteristics and the processing constants for 1363 and 1301 stocks 
do not change. It was felt, nevertheless, that certain tests should be 
made for every feature, so the following procedure was evolved. 

Master positives of every reel of a release, and the accompanying 
cross-modulation tests are first printed on 1363 stock to a density of 
2.05. The reels of master positive are stored, but the cross-modula- 
tion test positives are detached and printed on 1301 stock to make 
dupe negatives. The test from 1A is printed at three negative densi- 
ties and the tests from the remaining reels are printed to a density of 
about 2.00. Cross-modulation prints at several densities are then 
made from each of the dupe negative cross-modulation tests; and, 
from these prints the optimal print density for each dupe negative is 
determined. Reel 1A gives us a three-point slope curve, a typical 
example of which is shown in Fig. 10. 

Theoretically, these dupe print optima should be the same as the 
optima of the prints from the original negative,* for the master posi- 
tives and dupe negatives were made according to the technic set 
forth in this paper. But, experience has shown that changing proc- 
essing conditions and new emulsion numbers, sometimes, cause a 
shift of dupe optima. The slope curve of Fig. 10 then may be used 
for making corrections. 

The correct value of dupe negative density for reel 1A may be 
found directly from Fig. 10, as is shown, but the slope only is used for 
correcting the remaining reels. To use the slope, assume a hypo- 
thetical case where the original optimum of reel 2B was 1.55. We will 
further assume that the dupe negative density was 1.90 and that its 
optimum was found to be 1.45. Fig. 10 shows that an increase of 0.08 

"These values are on file, having been found previously to determine the 
release print operating range. 


in dupe negative density will raise the optimum 0.10. By raising the 
dupe negative then to 1.98, we will be assured of a negative which 
will match the original. These values of corrected dupe negative 
densities are kept on file for use during the printing of a release. 
When a negative is damaged, the correct dupe negative density is 
obtained from the files, and a section of dupe negative printed from 
the stored master positive. Immediately after development, the 
duplicate negative may be used to replace the damaged section with 
the assurance that it will take the same printer light as the remainder 
of the negative. 

This process has been in use on all Warner, First National, and 
Vitaphone releases for the past four months and has given con- 
sistently good results. 

In conclusion, we would like to point out that no claim is made 
that the process herein developed represents the ultimate in photo- 
graphic sound duping, but we do feel that the method of treatment is 
new and useful in developing any duping process. 


R. C. A. Mfg. Co., Inc., RCA- Victor Division, "Instructions for Quality Con- 
trol of Variable- Area Sound-Tracks," April 25, 1938. 

BAKER, J. O., AND ROBINSON, D. H.: "Modulated High-Frequency as a 
Means of Determining Conditions for Optimal Processing," J. Soc. Mot. Pict. 
Eng., XXX (Jan., 1938). p. 3. 

BAKER, J. O. : "Recording Tests on Some Recent High-Resolution Experimen- 
tal Emulsions," J. Soc. Mot. Pict. Eng., XXX (Jan., 1938), p. 18. 

BAKER, J. O. : "Processing of Ultraviolet Recording on Panchromatic Films," 
/. Soc. Mot. Pict. Eng., XXXI (July, 1938), p. 28. 

BLANEY, A. C., AND BEST, G. M.: "Latest Developments in Variable- Area 
Processing," /. Soc. Mot. Pict. Eng., XXXII (March, 1939), p. 237. 


Summary. Release print laboratories need some sort of device to check quickly 
sound-track center-line positions on release prints. No satisfactory device was avail- 
able, so a measuring machine was developed that indicates track position directly on a 
dial indicator. 

Since the advent of sound it has been necessary to measure the 
position of the sound-track on both positive and negative films, and 
for these measurements two classes of devices have been evolved. 
One group of instruments consists of standard microscopes to which 
film-holding devices and micrometer oculars or micrometer stages 
have been added. The other group of instruments consists of pro- 
jection devices which project enlarged images of the sound-track area 
on calibrated screens. 1 

Both these groups of instruments serve well the needs for which 
they were built. However, neither type is suitable for use in release 
print laboratories where a large number of sound-tracks must be 
checked rapidly by non-technical operators. 

The projection type of device requires a darkened room, occupies 
quite an area, and requires some mental arithmetic to arrive at the 
center-line measurement. For the quick checking of one print in 
many reels, it also consumes too much time in threading. 

The microscope types, while easy to thread, are too slow in opera- 
tion. With them, the procedure is somewhat as follows: A reading 
of the micrometer screw that moves the film-holding mechanism is 
taken when the edge of the film is beneath the cross-hair of the micro- 
scope eyepiece. Readings are then taken with each of the two bias 
lines, in the case of bilateral variable- width recording, successively 
under the cross-hair. Two readings are thus obtained one, the 
distance from the film edge to the inner bias line, which we will call A ; 

* Presented at the 1939 Fall Meeting at New York, N. Y. ; received October 
5, 1939. 

** Ace Film Laboratories. Inc., Brooklyn, N. Y. 




the other, the distance to the bias line nearest the sprocket hole, 
which we call B. The center-line will then be 


This bit of mental arithmetic is slow and fatiguing if continued all 

Since no existing device fulfilled the needs of our laboratory, we 
set out to design a measuring device that would be quick in thread- 
ing, occupy very little space, be portable, and require no calculation 
to arrive at the sound-track position. 

FIG. 1. Schematic arrangement of device. 

Fig. 1 is a diagram of the construction. Light from a small lamp 
shines through the film which is held by a spring parallel against 
the edge of a slightly curved gate which slides in Vee slides in a direc- 
tion perpendicular to the film direction. Motion is imparted by the 
lever 1, and the position of the film-gate is indicated by the one-ten- 
thousandth dial indicator 2. Directly above the film-gate, standard 
microscope optics have been mounted, which consist of a 32-mm 
objective lens and a 10-power Huygen's eyepiece. However, the 
usual cross-hair has been replaced with two parallel hair devices 
which slide in Vee slides. The lever 3 causes the hairs to move always 
in a direction parallel to each other, and equidistant from the optical 



Jan., 1940] 



center of the instrument. Fig. 2 is a general view of the completed 
device, which shows the film-holding device, the light switch, and 

FIG. 4. General side view. 

one of the operating levers. Fig. 3 shows both operating levers, the 
eyepiece, and the center-line indicating dial. Fig. 4 is a general side 

The operation is as follows: Film is 
placed in the gate. Levers 1 and 3 are 
moved until the two parallel hairs are 
each over a bias line, or over correspond- 
ing peaks, where there is modulation 
(Fig. 5). The center-line is then read on 
the dial indicator which has been cali- 
brated so that O on the dial corresponds 
to a track center-line position of 243 mils 
from film edge. Two markings, a plus 
and a minus, have been placed above the 
operating lever 1. They are useful in 
telling the sign of the deviation when 
the dial indicator has revolved more than one-half revolution in 
either direction. 

FIG. 5. View through 
eyepiece showing variable- 
area track and parallel 


Film may be inserted and the center-line read in ten seconds, and 
the reading may be duplicated to two ten-thousandths of an inch by 
different observers. 

In a release print laboratory we have no need for figures of per- 
centage modulation, but it is entirely possible to calibrate lever 2 
directly in percentage modulation, for the separation of the hairs is 
proportional to the amplitude. 

Two of these instruments have been in use for more than a year 
and are found to be very useful and accurate. They have needed 
no re-calibration, for the accuracy is dependent only upon the dial 


1 BEST, GERALD M.: "A Sound-Track Projection Microscope," J. Soc. Mot. 
Pict. Eng., XXXIII (Aug., 1939), p. 198. 


MR. PALMER: When you find that the center-line of the sound-track is not in 
the right place, what do you do about it? 

MR. ROBERTS: We adjust our printers. By printer adjustment we can even 
compensate for negatives which have misplaced sound-tracks. This device is used 
to control printing machines. A certain percentage of the output from each 
printing machine is measured, and a chart kept for each machine. When we ob- 
serve a drift as indicated by the chart, we adjust the printer. 

The machine has been calibrated so that the zero point on the dial represents a 
center-line of 243 mils from the film edge. The operating lever has been marked 
with plus and minus, so that the operator will know the sign of the deviation when 
the dial revolves more than half a revolution. 

DR. CARVER : Have you made any survey of the release prints to find what the 
general deviation is, as the theater man sees it? 

MR. ROBERTS: I am not acquainted with products other than the Warner re- 
leases. We keep our prints within =*=3 mils' maximum deviation. The average is 
probably better than that. 

Perhaps such a device would be useful in checking a large number of prints in 
an exchange, where, for a very rapid survey, it could be placed between rewinders. 


Summary. In view of its bearing on the design of ground-noise-reduction 
systems, a study was undertaken to determine how sudden or rapid are the increases 
in amplitude of the speech sounds that must be recorded in dialog. A large number 
of oscillograms were taken, a number of which are reproduced herewith. 

The most important observation is that the human voice can start several of the 
vowel sounds in such a way that the first wave is from 40 to SO per cent of the final 
amplitude, or, in other words, with a suddenness comparable to that of keying in an 
oscillator. However, this is rare, being for all practical purposes confined to a few of 
the more open vowel sounds, when not preceded by any consonant, and true only of 
certain individuals, depending on the manner of releasing the breath. Progressive 
build-up at rates wJiich would carry the modulation from zero to 100 per cent in 0.05 
second are frequent, while the great majority of syllables start more gradually than this. 

There are several types of apparatus which operate in response to 
the presence or the amplitude of an audio -frequency voltage or cur- 
rent. Telephone companies employ such devices for switching pur- 
poses, but the applications which are of principal interest to motion 
picture engineers are compressors and ground-noise-reduction sys- 
tems. The current for operating these devices is derived from 
rectified audio-frequency currents, and it is necessary, in order to 
prevent disturbing noises, that the rectified current be filtered, which 
is just another way of saying that the current is not permitted to 
make sudden jumps in value. Thus there is a limit to how closely 
the operating current can follow fluctuations in amplitude of the 
audio-frequency current. Failure of the current to fall quickly 
when there is a decrease in amplitude does not in general have any 
objectionable results, but failure to rise promptly when the audio- 
frequency amplitude increases causes distortion. In the case of a 
compressor, the initial waves are not cut down in amplitude as much 
as those immediately following, and an unnatural effect may result. 
If the compressor is depended upon to prevent overloading, it may 

* Presented at the 1939 Fall Meeting at New York; received October 29, 
** RCA Manufacturing Co., Camden, N. J. 




fail to act quickly enough, with a resulting quality loss, or a wax 
record may be overcut and ruined. In the case of ground-noise- 
reduction systems as applied to photographic sound recording, too 
slow opening of the masking device causes clipping of the tops of the 
initial waves. The industry has tolerated these faults, and efforts 
have been made to minimize them, but they are still with us. 

Obviously if the audio waves increase in amplitude only gradually, 
the small time-lag in response of the control current will have neg- 
ligible effect. In other words, it will, throughout the crescendo, 
have very nearly the value that it should have in view of the ampli- 
tude of the audio waves at that instant. The more rapid the in- 
crease in audio amplitude, the farther will the control current fall 
below the desired value. This is illustrated in Fig. 1. 


FIG. 1. Aggravation of wave-top clipping when rise is rapid. 

Ground-noise-reduction systems are designed Lo provide a margin 
of excess opening, in the form of an initial opening, which is main- 
tained even when the audio-frequency modulation is zero, and, in 
addition, a margin of excess gain is provided which in the steady 
state causes the masking to clear the tops of the waves by a generous 
percentage. It is these margins which protect the system from 
continual overloading or clipping at almost every beginning of a 
word or syllable, but, on the other hand, large margins defeat the 
purpose of the ground-noise-reduction system. 

A number of expedients have been proposed for speeding up the 
action of ground-noise-reduction systems. All these involve more 
or less complication, and no matter how fast the system may be made 
(within practical limits) there is always a possible rate of build-up of 
the audio-frequency waves such that the margin will be insufficient 
to prevent clipping. It is therefore pertinent to ask, "What is the 
steepest rate of build-up that will normally be encountered in sound 
recording?" and, second, "How often will these very rapid build-ups 


occur?" The studies of which this paper is a report, were undertaken 
for the purpose of answering the first question and extended in the 
hope of throwing some light on the second. 

It is well known that when there is considerable reverberation, 
maximum amplitude is not reached until after an appreciable interval 
after the start of the original sound. Since music is recorded in 
fairly "live" enclosures, it may be assumed that rates of build-up 
rapid enough to cause difficulties will not often be encountered in 
music recording. In "sound effects" almost anything might be ex- 
pected, and it hardly seems reasonable to attempt to anticipate 
what demands may be imposed on recording systems. It is probable 
that, at least in the case of the louder sounds, more or less reverbera- 
tion may in general be looked for. The major part of sound-on-film 
recordings is talk, and is recorded in fairly dead sets. Thus the most 
important part of the problem is concerned with how speech sounds 
start. An obvious approach would be to examine a large amount of 
recorded dialog. Such a study, if carried out on an adequate scale, 
affords the best answer to the question of frequency of occurrence of 
very sudden beginnings. Of more scientific and general interest, 
however, is a study in which all the sounds examined are identified, 
and several voices are used for each sound. 

The method chosen consisted hi producing a trace of the wave- 
shape on a cathode-ray oscillograph and photographing the screen 
with a film-pack camera. Three oscillograms could be photographed 
on each film. Mr. S. Read, Jr., proposed and worked out for us an 
arrangement for controlling the sweep of the cathode-ray beam 
which turned out to be very useful. When the circuit is set for an 
oscillogram, the beam is just off the screen on the left. Pressing a 
key releases the sweep and the spot travels at a very slow rate, of 
the order of an inch in three seconds. An extremely small audio- 
frequency disturbance serves to trip a thyratron which suddenly 
changes the bias on the grid of a pentode through which a condenser 
is discharged and the sweep then proceeds at a speed of 10 inches per 
second. The speaker watches the screen and upon seeing the spot 
crawl out onto the screen, makes the desired sound, which sound is 
recorded at the full sweep velocity. No waves can be lost by this 
method, although conceivably, if the release were not quick enough, 
the first waves might be compressed on the tune-scale. The release 
time is of the order of a few microseconds and the oscillograms do 
not show any evidence of failure to record all of the sounds. The 


slow travel of a light-spot causes the first part of the trace to appear 
very heavy, due to spread of exposure in the film. In several of the 
sounds, a period of very low modulation (too small to appear on the 
oscillogram) is indicated by the sudden narrowing of the horizontal 
trace. The beam became slightly defocused toward the edge of the 

r* *i>JLM 


\\\\ Y ; ]t 

1 mini 

FIG. 2. Examples of oscillograms. 

screen so that detail is lost here in the oscillograms, but this did not 
defeat the purpose of showing the starting characteristics. We are 
concerned primarily with the envelope of the wave, so no effort was 
made to correct the defocusing. Consideration was given at first 
to showing only the envelope rather than a true oscillogram, but this 
would fail to show some factors which might turn out to be of con- 
cern. For example, if a rapidly rising envelope is produced by com- 


paratively few waves, the action on the rectifier will be materially 
different from the case in which the envelope widens at the same rate, 
but there are shorter intervals between the peaks. It will be noticed, 
for example, that the oscillograms by E. K. show a remarkably wide 
spacing between sharp peaks, although the voice fundamental in 
this case was not much lower than that of the others. This was at 

FIG. 2 A . Tracings of oscillograms of Fig. 2. 

first attributed to some error in the speed of the sweep, but further 
tests showed the same characteristic. If there was any departure 
from the correct speed during any of the tests, it was small enough 
to be of no significance. 

Fig. 2 shows a number of the oscillograms. Those who have tried 
photographing cathode-ray oscillograms will understand the difficulty 
of getting a picture which is entirely satisfactory for copying by a 
halftone process. Although our oscillograms were entirely legible, 

48 R. O. DREW AND E. W. KELLOGG tf. S. M. P. E. 


. |^ 

FIG. 3. Tracings of oscillograms of speech sounds. 

we feared that many of them would not make satisfactory engravings. 
Some of the peaks, for example, were so faint that they might be 
lost. We decided that a tracing, with a dot at each peak, would 
show about everything that had an important bearing on the present 
study. Therefore we had a set of such tracings made, and are de- 
pending on these to show the growth characteristics of the various 
sounds. Fig. 2A shows the tracing corresponding to each of the 
oscillograms shown. In the tracings the end of the heavy line indi- 
cates where the circuit tripped and the spot moved at full velocity. 
Dots substantially along the axis before the main sound indicate the 
existence of a small preliminary modulation, as for example, a con- 
sonant. It is obviously impossible as well as unimportant for our 





E L. "A" CE -6.) 00(AS i~ MOOO) PUT 

FIG. 3 (Continued). Tracings of oscillograms of speech sounds. 

purpose, to show the exact amplitude or wave peak positions in this 
low modulation. The entire set of tracings is shown in Fig. 3. 

The speaker faced the ribbon microphone at a distance of about 
two feet, in a heavily damped room. Before making the oscillo- 
grams here shown, the effect of varying the distance from the speaker 
to the microphone was tried. No significant change was noted. 
Tests were made also with the microphone from three to four feet 
from the speaker and a reflecting surface about the same distance 

50 R. O. DREW AND E. W. KELLOGG (J. S. M. P. E. 



YOU 'V" ox WISE 

FIG. 3 (Continued). Tracings of oscillograms of speech sounds. 

beyond the microphone. This was for the purpose of answering the 
question whether such echoes as come from the walls of a motion 
picture set would materially alter the story of how rapidly sounds 
start. It is assumed that the set would be so constructed that there 
would be a negligible amount of multiple reflection. The effect of 
the single echo produced as described above was, as might be antici- 
pated, quite small and in no wise prevented the sounds from building 
up to a large fraction of their final amplitude, substantially as quickly 
as without the reflection. 

As has already been indicated, there was doubt in the minds of 
the writers whether the human voice mechanism is capable of build- 
ing up sounds so rapidly as to reach substantially full amplitude in 




S* s. 158 

ACCENTED ! - Z0 .|;i!T.7.V.'.7 ZJ Hfii ^ l " KH^SVSS ZSS 

ACCENTED - J30 I.'.'.'.'.. 32 1 _,..---^ J2J 


//ft J/ 

FIG. 3 (Continued). Tracings of oscillograms of speech sounds. 

one or two cycles. Since a number of resonators are involved, it 
might be expected that the voice would have the characteristic which 
is usual in oscillators, namely, that a considerable number of cycles 
is required to reach full amplitude. This depends on the ratio of 
feedback to damping, and the writers had little knowledge of these 
factors. It did not take many tests to show that it is quite possible 
to reach practically full amplitude even on the first wave. There- 
fore cases will occur in which clipping of the first wave can not be 
avoided by a ground-noise system of any speed within the range of 

52 R. O. DREW AND E. W. KELLOGG tf. S. M. P. E. 

practicable application for ground-noise reduction. Such clippings 
must thus be tolerated unless an anticipation system can be employed. 

The next question is, how often will this very sudden burst of 
sound be encountered. We had hoped that our studies would give 
some indication on this point but soon discovered that this would be 
a statistical matter not readily subject to direct experiment. There- 
fore this phase of the problem must be deferred, and we confine our- 
selves to an examination of the relative tendency of various word 
sounds to produce these rapid crescendos and to study the nature of 
the increase in amplitude. 

The oscillograms may be classified into : 

(1) Those which increase gradually from zero to substantially full amplitude. 

(2) Those which increase in the same continuous manner but more rapidly. 
(5) Those which jump suddenly to an appreciable fraction of maximum 

amplitude and thereafter rise slowly. 

(4) Those which follow the initial jump by a more rapid increase to full 

(5) Those which show fifty per cent or more of the maximum amplitude on 
the first or perhaps the second wave, thus giving an envelope with a large 

There may also be, in combination with any of the sounds described 
above, a preliminary modulation of low amplitude, ordinarily the 
result of some consonant sounds preceding the vowel. This pre- 
liminary low-amplitude modulation is an important factor in design 
of ground-noise-reduction systems, hi that it can be used to start the 
shutter opening, and this will materially reduce the likelihood of 
clipping the larger-amplitude waves. 

Inspection of the oscillograms shows that the starting characteristic 
is far more dependent on the person talking than on the specific word 
sound. For example, R. D. produced shoulder-type starts in a very 
large fraction of the sounds, while H. B. produced only one. Neither 
are successive utterances of what is supposed to be the same vowel 
sound consistently the same in envelope nature when spoken by the 
same person. 

The gain in the system is substantially constant throughout the 
series, so that the differences in final amplitude are almost entirely 
a question of how loud the person talked. If a person talked con- 
sistently, a fair comparison could be made between the amplitudes 
which are likely to appear in various vowel sounds. The larger 
amplitudes are produced only by the vowel sounds. Most conso- 


nants are relatively very weak. The semi-vowels, or sustained con- 
sonant sounds: /, m, n, z, j, give considerably larger amplitudes than 
most other consonants, but are still far below the vowels. R is 
intermediate. On rare occasions 5 may produce large amplitudes, 
particularly if high frequencies are exaggerated for recordings, as is 
frequently done, but this is so unusual as to be of no practical im- 
portance for the main purpose of this study. An extensive study of 
the subject has been made by C. F. Sacia. 1 The figures in Table I 
on vowel sounds are taken from Dr. Sacia's paper. 


Relative Power and Peak Factors in Vowels 

Male Voices Female Voices 

Peak Peak 

Speech Sounds Power Factor Power Factor 

OO (Tool) 27 2.6 41 2.8 

OO (Hood) 33 4.0 40 3.1 

6 (Oh) 33 4.1 44 3.4 

Awe 37 4.5 50 3.3 

Up 29 4.6 38 3.9 

Ah 50 4.2 48 3.6 

Add 44 5.4 39 4.7 

6 (ever) 26 5.6 31 3.8 

A (Aim) 22 5.3 30 4.5 

K (/*) 25 4.1 32 3.8 

5 (eel) 23 4.7 23 2.6 

In Fig. 3 the vowel sounds have been arranged roughly in order 
of the successive mouth positions, from wide open to nearly closed. 
This is likewise substantially the order which they would take if 
arranged in order of decreasing tendency to produce large amplitude. 
Abrupt starts or "shoulder-type" envelopes are somewhat more 
common hi the open vowels. 

We had rather expected that preceding a vowel by one of the ex- 
plosive consonants, p, t, k, would be especially likely to produce an 
abrupt start. This is not the case. It appears that the likelihood of 
large shoulders in sounds beginning with p, c, and / is about the same 
as for the corresponding vowel sounds when not preceded by any 
consonant. The most sudden attack seems to be produced by build- 
ing up some air-pressure in the lungs, with the epiglottis closed, and 
releasing it suddenly, which has the effect of beginning the sound 
with something like a grunt. This is widely done in greater or less 
degree and is probably a characteristic of the individual. It can, of 


course, be done with any vowel. It probably has something to do 
with whether the person has the habit of talking with the throat re- 
laxed or under certain amount of tension, but in the cases of one or 
two speakers deliberate effort to control the type of start had very 
little effect. It should be remarked that whatever habits of voice 

FIG. 4. Effect of turning the microphone 180 degrees. 

the speakers may have had, the tendency to produce shoulder-type 
starts is not associated with any observable hardness of voice or any 
jerky quality, such as might perhaps have been expected. Only 
one woman's voice is represented, the oscillograms marked AM. 
All the rest are of men's voices. Preceding the vowel by almost any 
consonant has the effect of promoting a gradual beginning. The 
consonant not only provides an initial or warning low-amplitude 


modulation, but is conducive to a more gradual build-up in the vowel 
sound itself. 

In some preliminary observations, it appeared that the amount of 
shoulder was little affected by whether the sound was accented or 
not, the main effect of accent being to cause the vowel to continue its 
crescendo to a higher final value. Oscillograms taken subsequently 
in which the same sound was spoken successively without and with 
accent did not altogether bear out this observation, the principal 
effect of accent being to raise the amplitude throughout the whole 
range, or, in other words, to magnify the envelope in the vertical 
direction. The accent comparisons are shown in Fig. 4. 

One of the interesting things to be observed in these oscillograms 
is the consistency with which upward deflections exceed the down- 
ward deflections, whenever there is any difference. That this was 
not due to any distortion in the equipment is proved by simply turn- 
ing the microphone 180 degrees. In Fig. 4 all the oscillograms (with 
the exception of Nos. 181 and 183) made with the microphone in 
the regular position are shown above, and just below each of these 
is an oscillogram made immediately afterward, with the microphone 
reversed. This characteristic of speech sounds was pointed out by 
S. Read, Jr., in connection with his discussion of the "Neon Volume 
Indicator" 2 and its applications, and has recently been discussed by 
J. L. Hathaway. 3 It has further been observed by Mr. Read and 
other engineers in the RCA laboratories that if clipping does occur, 
it is less objectionable when it is the short peaks that are clipped 
than if the high peaks are clipped. 

Both these facts are important factors in ground-noise-reduction 
systems. If the ground-noise-reduction system is designed to work 
on the negative half -waves, not only is less shutter movement (or its 
equivalent) needed to avoid clipping, but the damage to quality if 
clipping occurs is less serious than if the opposite polarity is used. 

Such observations and generalizations as we have so far reported 
apply to both the pressure and rarifaction halves of the waves. In 
general the positive and negative envelopes are of about the same 
shape (as if the same curve were simply plotted to a different scale). 
When there is any difference in shape at all, it is almost without ex- 
ception in the direction of slower build-up for the negative half- 
waves. A difference in shape which appears in a considerable number 
of cases is that the negative build-up is delayed with respect to the 
positive, as, for example, oscillograms Nos. 57, 39, 5, 22, of the U and 

56 R. O. DREW AND E. W. KELLOGG [J. S. M. P. E. 

F sounds. When a sound of this type is produced, a ground-noise 
system so designed that it has to clear only the negative peaks, is 
given the benefit of a warning modulation of small amplitude, before 
the major increase begins. 


1 SACIA, C. F. : "Speech Powers and Energy," Bell Syst. Tech. J., 4 (April, 1925), 
p. 627. 

2 READ, S., JR.: "A Neon Type Volume Indicator," J. Soc. Mot. Pict. Eng., 
XXVIH (June, 1937), p. 633. 

1 HATHAWAY, J. L.: "Microphone Polarity and Overmodulation," Electronics 
(Oct., 1939), p. 28. 


MR. OFFENHAUSER: What kind of microphone was used, and what was the 
distance from the microphone to the speakers? 

MR. KELLOGG : The person sat about two feet from the microphone in a heavily 
damped room. We tried some tests with reflectors, but with no appreciable re- 
sult. The effect of the reflector was not to produce anything approaching real 

MR. BERGER: What was the maximum relative change on the steepest curve 
found, between the point of where you could just about see the change and where 
it showed up as a rather large change? 

MR. KELLOGG: We have only the oscillograms to go by on that. The pre- 
liminary vibrations were probably between 20 and 30 db below full modulation. 

MR. BERGER: Did the peak show up in the period of cycle of the wave? On 
some of the curves it looked as if it might be 200 cycles. 

MR. KELLOGG : The little spots on the diagrams that mark the tips of the waves 
show about what happened. There was a little disturbance, followed sometimes 
by only one small reverse wave, and then the pressure would build up to practi- 
cally full amplitude in the positive direction. In other cases, there would be a 
couple of waves. There were one or two cases where there was hardly any warn- 
ing, and the first wave went as high as any of the succeeding waves. 

MR. BERGER : Roughly speaking, would you say it rose up to its full magnitude, 
in the fastest one you found, in a hundredth of a second or less? 

MR. KELLOGG: If the first wave shows full amplitude the pressure must build 
up from zero in about */4 cycle, or in about VMO second in the case of a 125-cycle 
wave. Of course, the higher-frequency components contribute to building up 
the peak, so that it is difficult to say just how quickly it is reached in the case of 
the most sudden build-ups. It should be remembered that starts of this kind are 
rather rare. I would like Mr. Drew to tell us his experience in trying to control 
the suddenness of beginning. 

MR. DREW: It seems almost impossible for any given voice to deviate volun- 
tarily to a slow build-up from what might be a very rapid build-up for that voice. 
For example, my voice has a very steep shoulder on almost all the vowel sounds. 
Mr. Kellogg's voice and some others that are about the same in frequency range 
seem to be very fast on the first wave. Other voices started more slowly, and the 


initial waves were cone-shaped from zero to full amplitude. Many vain attempts 
were made by speakers with steep "shoulder type" voices to produce slow cone- 
shape build-ups. 

MR. OFFENHAUSER: How would the rate of build-up of a single piano tone 
compare with the rate of build-up of certain of the speech tones you have shown? 
Have you noted any similarities? 

MR. KELLOGG: We did not make any tests with a piano. Percussion instru- 
ments no doubt build up very fast, but I do not think the piano does come to full 
amplitude right away. Some of the percussion instruments no doubt would; 
but in the case of the piano the sounding board is large and heavy and there are 
numerous elements besides the strings which may resonate. As pointed out in 
the paper, musical instruments would not be played in dead rooms. 

MR. Ross: In voice culture four differing groups of consonant vowel for- 
mations are recognized. The first are the labials, as, for example, pah, bah, mah, 
wah, wherein the lips are involved. The second are the palatals, such as tah, dah, 
nah, lah, and rah, wherein the tongue is involved. The third are the labial pala- 
tals, as, for example, fah and wah, formed by both the lips and palate; and the 
last group are the gutterals, such as kah and gah. In the last group, the con- 
sonant formation begins at the vocal cords and accounts for the gradual, instead 
of abrupt, increase in amplitude of the sound produced, as shown by the first 
record referred to. The most explosive sounds are those produced by singing or 
speaking the pure vowels, as, for example, ah, aye, ee, eye, oh, or you. 

MR. KELLOGG : We covered about all the types of sound we could think of, in- 
cluding those you mentioned. The examples are not sufficient in number, of 
course, to give anything of value statistically. It appeared from our tests that 
beginning a word with an unvoiced stopped consonant is not conducive to any 
greater suddenness than that which results if you simply start with the epiglottis 
closed and suddenly release your breath, but as Mr. Drew said, it does not seem 
possible to control the nature of the start entirely. 

One conclusion that we seemed to be able to draw from our tests, and from 
the fact that so many sounds can begin suddenly, is that the human vocal cords 
are not like a cornet, where a return wave from an air-column would control 
the vocal cords and thereby establish when the next pulse comes. If a strong 
resonance of some kind were involved, such sudden beginnings would not be 
possible. The voice mechanism seems more like a relaxation oscillator that 
does not carry over any energy from one cycle to the next. 

MR. Ross: That is true. However when producing the pure vowel sounds, 
the maximum amplitude of sound is produced immediately without waiting for 
shading thereof by a preceding consonant. Therefore the initial recording will be 
of maximum amplitude. 

MR. KELLOGG : That did not seem to increase the suddenness rather the re- 
verse. The voice sound with the lips closed produced a little preliminary modula- 
tion, enough for starting the operation of voice-controlled devices; and then the 
opening of the passage takes place over a period of several cycles. The increase is 
something like the action of swell in an organ the pipes are sounding but the 
full volume outside is reached only gradually. I should like Mr. Reed to tell us 
something of the first observations of the consistent asymmetry of voice waves. 


Mr. READ: About six years ago we were working with a neon volume indi- 
cator where the tubes flashed on only one-half of the wave. In the particular 
model, we had not taken the precaution to have all the tubes flash on the same 
half of the audio wave. On some speech sounds the lamps failed to break down in 
proper sequence, as they had done when calibrated with a sine-wave signal. 
We found that the lamps that were skipping were ones that were actuated by the 
portion of the sound-wave due to reduced pressure. Most of them worked on one 
polarity, while a scattered few were controlled by the opposite polarity. At that 
time we made extensive tests which proved that the effect was not due to any 
distortion in the system. We were using bidirectional microphones, so one test 
consisted in turning the mike around and talking into the other side. This re- 
sulted in different lamps flashing out of sequence. 

Apparently such sounds consist of a series of impulses or shocks, the effect of 
each dying out before the next impulse occurs. It seems to come on the pressure- 
wave with most individuals; and, if you think about it, that is the easiest type of 
wave to generate. We attempted to generate a rarefaction trying to suck the 
sound in is what it amounts to. Not many persons, it seems, are able to do it. 
It is reported that a few actors in Hollywood produce sound-waves where the 
instantaneous reductions in pressure exceed the instantaneous increases, which is, 
in other words, the reverse of normal speech. 


Summary. The contention that a linear recording and reproducing system rep- 
resents the ideal, and that sound handled by such a system will be exactly represented, 
is not borne out by experience. Systems have been built which meet this requirement 
within limits that are not detectable by the ear and yet these systems do not reproduce 
sound as it actually is produced. In many cases a definite non-linear response curve 
is provided to compensate for some factor that is not covered by the above contention. 
It is the author's thesis that this discrepancy is due to the ear sensitivity to frequencies 
as a function of loudness. 

Using the ear-sensitivity curves presented by Fletcher and Munson of the Bell 
Telephone Laboratories (which have been verified by other observers} it is shown how 
the ear introduces frequency distortion to a linear system when the sound is reproduced 
at a level other than the level at which it is produced. It is shown how a sound repro- 
duced above the incident sound-level introduces excessive low frequencies. The case 
for a sound reproduced at a lower level is also examined and the conclusion is drawn 
that this case accentuates the high frequencies. 

It is further shown that the possibility of correcting for the limited volume range of 
all sound systems may lie in the type of amplifier response curve. 

A description is given of three methods used to achieve the desired amplifier char- 
acteristics: (Jf) a mechanical method, (2) a linear-non-linear system, and (5) a 
selective by-pass system. Circuits are given and the important operating points of 
each are discussed. The objections to each system are also given. 

Further, a brief summary, with diagrams, describes the various set-ups used to 
record with these amplifiers. This covers work for radio, disk record, and sound-film. 

In the past the problem of recording and reproducing sound has 
been approached primarily from the standpoint of frequency and 
harmonic distortions inherent in the mechanical, electric, and chemi- 
cal elements of the various systems employed. So much progress 
has been made toward the elimination of these distortions that, to- 
day, it is quite correct to say that frequency and harmonic distortion 
can be held within limits that are not objectionable. Now that these 
system distortions have been corrected, it seems that a detailed exami- 
nation of the effect of reproduction volume should be made. 

* Presented at the 1939 Fall Meeting at New York, N. Y.; received August 
28, 1939. 

** Case School of Applied Science, Cleveland, Ohio. 




[J. S. M. P. E. 

Although data have been compiled by Fletcher and Munson of the 
Bell Laboratories, and although many have noted the effects of 
volume on quality, there seems to have been no detailed analysis 
made of this effect. 

This paper proposes to analyze this effect, and to indicate methods 
of correcting the distortions produced. 


Ear-Sensitivity Curves. Fletcher and Munson have shown that the 
ear does not respond to all frequencies with the same sensitivity. In 

mo 10000 / 

FIG. 1. Ear-sensitivity curves. 

addition, they have shown that the ear varies in sensitivity to fre- 
quencies as a function of the loudness of the incident signal. 

Fig. 1 shows the ear-sensitivity curves of Fletcher and Munson. 
As these curves are not straight lines, they show what, in an amplifier, 
is termed frequency distortion. However, this fact is not the salient 
point of the curves. 

The important point is that the slopes of the curves at correspond- 
ing frequencies are not the same for all curves. If the slope were the 
same then the ear would hear all signals, at all loudnesses, in the same 
proportion of energy at the various frequencies. If this were true 
there would be no distortion to the ear. 


Examination of these curves shows that the condition of equal slope 
is substantially met at frequencies above 1200 cycles. That this is 
fortunate will be shown when the amplifier design is discussed. 

Implication of Curves. The implication of these curves can best be 
shown by an illustration. Consider a system carrying a 100-cycle note 
and a 1000-cycle note, each containing the same energy represented 
by 80 db. Fig. 1 shows that the ear will hear the 1000-cycle note as an 
80-db note while the 100-cycle note will be heard as an 83-db note. 

If the system gain is lowered 40 db, the intensities of the two notes 
are actually 40 db but are heard by the ear as sounds of 20-db differ- 

Distortion Produced as Function of Curves. To illustrate the re- 
sult of this effect, assume that it is desired to record a sound of uni- 
form intensity at all frequencies. Also assume this sound occurs at 
50 db. 

To an observer on the set the various frequency components of this 
sound will actuate his ear in a manner shown by the ear-sensitivity 
curve for 50 db. However, the electrical system recording this sound 
is built to receive uniform electrical energy. Therefore, the electrical 
system does not record the sound in the same manner as heard by 
the observer. And due to the varying ear sensitivity of the observer 
as the loudness varies, the amount of distortion between the ob- 
served sound and the recorded sound varies. 

Condition for No Distortion. From this it follows that the con- 
dition for no distortion is that reproduction should occur at the same 
level as the incident sound. When this is done the reproduced sound 
will be heard by the ear of the observer in the same manner as the 
observer would hear the sound as it occurred on the set. 

If this is not done, distortion is introduced. 

Effect of Reproduction Level above Incident Level. If reproduction 
takes place at a level above the incident level the effect is to accentuate 
the low frequencies. This is shown by Fig. 2. The curves are found 
in this manner : 

Assume a constant-energy sound occurs at 20 db and is reproduced 
at successively higher levels. At 20 db the ear has one response. The 
difference between each higher response curve and the 20-db response 
curve, at various frequencies, is the error introduced. The curve thus 
found from Fig. 1 is the sensitivity difference. The conjugate of 
these curves gives the apparent added energy at the low frequencies. 
These are shown by Fig. 2. 



(J. S. M. P. E. 

Effect of Reproduction Level below Incident Level. By similar anal- 
ysis Fig. 3 is found. Here it is shown that reproduction at a lower 
level than the incident level produces an exaggeration of the high 

Condition of Varying Energy Signals. The static condition has 
been discussed. Examination of the varying energy case will now be 
made. If a sound begins at a given level and is reproduced at another 
level, a certain error curve is determined. However, if the sound 
increases in level, the reproduced sound increases in the same amount, 
but the error curve is different. 

Ar Hum* Lfvl. 

toco 10000 f 

FIG. 2. Effect of reproduction level above incident level. 

Specifically, a sound of 20 db reproduced at 30 db will generate one 
error curve. A 5-db rise in incident sound will raise it to 25 db and 
the reproduced sound to 35 db. The error curve for the 20-30 db 
combination is different from the error curve for the 25-35 db combi- 

Therefore, it is seen that an infinite number of curve combinations 
occurs for the infinite possible combinations. And it is also seen 
that a static response curve will not correct this error. 

Effect of System Range. The above example considered the in- 
cident sound as continually increasing without limit. However, all 
recording systems have a maximum level that can be recorded. When 
this top point is reached the incident level continues to increase but 


the reproduced level remains at the maximum the system can handle. 
After this point is reached the error curves produced represent the 
difference between the ear-response curve at the top of the system 
range and the increasing incident sound-level. 

This points out the possibility that this type of error can be cor- 
rected by adjusting the response curves of the system, and thereby 
simulate the effect of increasing loudness without an actual increase 
in power output. 

Application of Correction. Correction for volume distortion is de- 
pendent on the system to which it is applied. Theoretically the 
following should be done: As will be pointed out later the actual 
correction can be done quite readily at the recording end. 

Radio. It is not to be assumed that all radio receivers are played 
at the same loudness, nor can it be assumed that any has a linear re- 
sponse. Therefore, it would seem that any correction should be ap- 
plied to the receiver proper. 

On the other hand, the average radio has a range of about 35 db. 
As there is no metering element that may be introduced at the re- 
ceiver to compensate for distortion due to system limitations this 
type of volume distortion should be corrected at the transmitter. 
The correction should compensate for the lower level of reproduction, 
which accentuates the high frequencies. 

It seems that from the above theory it may be concluded that the 
layman who uses all the bass obtainable with his tone control, and 
the layman who insists that acoustic labyrinths, baffles, and resonance 
chambers be added to his radio is more correct than is the engineer 
who insists on linear response at all levels. 

Records. Substantially the same may be said with respect to disk 
records as is said for radio except here the system range is about 
55 db. 

Sound-Film. Sound-film systems represent a different condition. 
In these it is necessary to correct for a reproduction level above the 
incident level, which accentuates the low frequencies. It is necessary 
also to correct for system limitations. 

Correction may be applied at the reproduction end but advantage 
should be taken of the fact that most reproduction occurs at approxi- 
mately the same level and this allows correction to be made at the 
recording end. An advantage gained by correction at the recording 
end is that film noise, occurring mostly in the high frequencies, is 
blanketed by the excessive highs introduced in recording. 



[J. S. M. P. E. 

Actual Correction. Fortunately it has been found possible to cor- 
rect at the recording end in all cases. This is a great advantage and 
is done on the assumption that approximate reproduction levels 
may be assumed. Investigation shows that radio reception occurs 
seemingly at all levels. One trouble introduced by radio is the 
limited frequency response of receivers. Although stations transmit 
a 10,000-cycle band a good commercial receiver will cover only 3500 

Conclusions. The major points that may be concluded from this 
discussion is that no static response curve for a system, because of 

epfeer of R*r*oi><jcTioM if 
BSLOW InciotHT LevcL 
(4OO8 SISNUL Ptrtooucio 
Lowe* Ltv* 

5000 10000 f 

FIG. 3. Effect of reproduction level below incident level. 

the varying ear response curves as a function of loudness, can record 
and reproduce a sound exactly. 

A point that should be mentioned is that this theory is based on 
the average ear response curves. Therefore, assuming that the 
system response may be continually adjusted, these adjustments will 
fit only one observer. However, it must be borne in mind that the 
ear-sensitivity curves could not be presented unless there is agreement 
between test subjects within at least 25 per cent. If this is true then 
correction will be 125 per cent for one observer and 75 per cent for 

If it is assumed that there are some ears that respond in a totally 
different fashion, then this correction will not please that observer. 


Another point is the case of specially trained ears, such as the 
musician. His response curves may fit the average within the 25 per 
cent limits, yet due to his education on musical sounds this correction 
may not please him. 

These points seem to be the limiting factors in the correctness of 
this interpretation of the effect of the ear response curves on a record- 
ing system. 


Part I of this paper has attempted to explain the origin of volume 
distortion and evaluate the magnitude of this distortion. 

The salient point indicated for correction of this distortion was the 
explanation showing that a static characteristic curve for a reproduc- 
ing system was not sufficient. The system must have a varying re- 
sponse curve. This response must vary as a function of loudness, or 
energy input to the system, and also as a function of the output 
loudness of the system. 

The curves that represent the characteristics of these systems can 
be found as indicated in Part I as the conjugate curves of the error 
curves shown in Figs. 2 and 3. This means that the curves of the 
system are not finite in number. For every condition of incident 
loudness compared to reproduced loudness there exist system curves 
that are different. 

This makes necessary the adoption of reproducing levels to work 
around and also the use of approximations for response curves. 
Therefore, although the general trend of the correction curves can be 
established, the final arbiter of the desired curves is the ear. 

With this in mind the work on achieving correction was conducted 
along three lines. All were thoroughly investigated and one adopted 
and now in use. 

Assumption. To form a basis upon which to design the requisite 
amplifiers it was necessary to determine the average loudness of 
reproduction of the various systems to which this correction is to be 
applied. When these levels are determined it is possible to determine 
the successive error curve for each loudness level above this average 
level and then correct between these curves by proper amplifier 

Two methods for determining these error curves present themselves. 
It is possible to assume that the reproducer level remains constant, 
and then draw the curves showing the error between this level and 

66 S. L. REICHES [J. S. M. P. E. 

the succeedingly increasing levels. Second, it is also possible to draw 
the curves between the assumed level of reproduction and the incident 
level producing this reproduction level, and then draw the error 
curves for each successive increment of input to output. 

Determination of Levels. A pure tone is sent over a line and is 
reproduced. A series of observers in an average size room adjust 
the level until it is adjusted to their impression of the loudness their 
receiver would be played. This loudness is measured and is used as 
the level for both radio and disk records. The same follows for film 
work. As it is necessary to determine only one level, by the assump- 
tion made above, the requisite curves can be drawn. 

To check these figures a set-up was made enabling a conversation 
at normal levels to be measured after the loudness has been set by 
several observers. 

Definition of Average Loudness. For the purpose of the problem 
there can be said to be three different loudnesses for each subject 
recorded. These may be termed (1) general loudness; (2) specific 
loudness; (5) instantaneous loudness. 

Any sound occurring over a period of time may be said to have 
these three factors: 

(1) General loudness is the average loudness of a sound over 
the duration of its occurrence. This is the average loudness referred 
to when an average level is denoted for a given sound. This is the 
loudness for which the equalizers are used. 

(2) Specific loudness is the average loudness of a sound for an 
interval of time much less than the duration of the sound. This is 
the loudness referred to when it is said the loudness of a sound varies 
between 30 and 60 db. 

At any interval of time less than the time for the general loudness 
but greater than the audio-frequency changes there is a specific 
loudness. This is the loudness variation that should be corrected 
by the amplifier. 

(5) Instantaneous loudness is the loudness that varies with the 
envelope of the audio-frequency. 

Amplifier Considerations. It was decided to build a standard 
amplifier so that control would be had over the high frequencies. 
If the high frequencies are to be varied it can be seen that the type 
of variation, whether an increase or decrease, must depend upon 
whether reproduction takes place above or below the incident level; 
that is, for a reproduction level below the incident level it is neces- 

Jan., 1940] 



sary to add low frequencies or subtract high frequencies. As the 
level rises the amount of lows added must decrease. When the 
reproduction level is above the incident level the correction which 
consists of an excess of high frequencies must also decrease. 

In terms of an amplifier that controls only the high frequencies 
this means that the louder the source becomes the more highs must 
be added for a low-reproduction level and subtracted for a high- 
reproduction level. 

This is self-explanatory with a lower reproduction level. It is 
desired to add highs as the level rises. Therefore, the highs can be 

FIG. 4. Mechanical response control. 

increased. However, if the highs are removed from the high- 
reproduction case the distortion will be exaggerated. This is ad- 
justed by the equalizer. The equalizer produces the average cor- 
rection curve which raises the high frequencies to a point above a 
linear response and the A. R. C. amplifier, as a function of the signal, 
will cause highs to be removed. 

If the A. R. C. amplifier is to swing about the average correction 
curve it will be necessary to have the A. R. C. change about a line 
other than a linear response; also this line would vary for each 
reproduction case. However, if the equalizer is adjusted to fit the 
low points of the signal, the A. R. C., varying from a linear curve, 
will fit all cases. 



[J. S. M. P. E. 

Mechanical System. This method takes advantage of relays that 
operate on small currents of the magnitude of 1 to 1.5 milliamperes. 
A group of fifteen relays was used. They operated as a function of 
signal magnitude. These relays are connected (Fig. 4) in conjunction 
with resistance A such that on being actuated a definite portion of this 
resistance, which is in a series with an equalizer network, is removed, 
and the resistance of the line from the network to ground is decreased. 
These steps were of the order of 1 db. 

The relays are operated in series with the plates of a tube operated 
in Class B. Thus the increase in signal increased the plate current, 
which in turn actuated the relays. The progressive operation of the 
relays is adjusted by means of resistors B. 


FIG. 5. Linear-non-linear combination. 

Condenser C is used to by-pass the audio in the plate line of the 
Class B tube. This is necessary as the audio induces a signal in the 
contact arms of the relay and is fed back to the main signal line. 

Although this system can be made to operate there are several 
objections to its use. The obvious objection is the number of parts 
which must be adjusted regularly. In operation the unit is very 
noisy and this noise appears electrically as clicks. These clicks may 
be suppressed to some extent, but they remain objectionable. 

In addition to this the usual action of a relay, opening at a current 
less than the closing value, causes some trouble. The relays used hi 
this test were sufficiently fast and were not troublesome in this re- 

Jan., 1940] 



spect. Because of these reasons, as soon as the unit was built and 
proof was had that the system would work to some extent, the work 
was discontinued. 

Linear-Non-Linear Combination. Another method tried was the 
combination of a linear and non-linear amplifier. Linear and non- 
linear refer to the ratio of signal input to signal output rather than 
frequency response. 

The linear amplifier was a typical voltage amplifier. The non- 
linear amplifier took advantage of changing the gain of a multigrid 
tube as a function of the bias on the one of the grids. This circuit is 
shown in Fig. 5. The bias of the multigrid tube is varied by the 

FIG. 6. Selective by-pass circuit. 

change in d-c voltage developed across R by the current from the diode 

The gain of the non-linear amplifier plus the action of the ex- 
pansion delay voltage, which may be added at A (Fig. 5), allows con- 
trol of the shape of the combined response curves as a function of 

This system works but has one very bad objection that is inherent 
in the non-linear amplifier. The expansion action, being a function 
of the grid bias of the expander tube, will occur without harmonic 
distortion only within quite narrow limits. And as the expansion 
required is quite high, distortion is quite bad. 

In addition to this another point enters, in that the d-c bias control 
voltage contains half-wave audio which feeds back to the main 
signal line on heavy signals. 

70 S. L. REICHES [J. S. M. P. E. 

The third objection to this method is that the expansion action of 
the non-linear amplifier increases the power delivered to the line. 
These sudden surges are impossible to catch with manual monitoring. 
A peak-limiting compressor can handle some of this. But at best 
this is an objectionable feature. 

Selectiie By-Passing. This method, which was finally adopted as 
the most desirable, consists of a system for by-passing selected fre- 
quencies as a function of loudness. Fig. 6 shows the circuit for this 

The voltage amplifier has substantially a linear frequency re- 
sponse. To this is added the by-pass line. The line consists of a con- 
denser in series with a variable resistance. This variable resistance is 
the plate-to-cathode resistance of a triode tube. 

The resistance of this tube is controlled by the d-c impressed on 
its grid, produced by a diode pair in full-wave rectification. The 
diodes are driven by a duo-triode with one section used as an inverter. 
The d-c voltage used is positive and is developed across R. The con- 
denser is for filtering action. The battery in series with the positive 
d-c voltage is of great importance. 

For a low-level signal a linear frequency response is desired. There- 
fore there should be no by-passing through the condenser and tube. 
This means that the tube should be blocking, and theoretically of 
infinite resistance. This is achieved by means of the battery. This 
battery swings the grid of the by-pass triode sufficiently negative to 
block the tube, and to block it enough for a given gain on the d-c 
source that no by-passing occurs until desired. 

As soon as the signal level increases, the d-c from the diodes in- 
creases and, with the same battery voltage, the effect on the grid is 
to make it less negative. This decrease in bias on the tube lowers its 
resistance and current is by-passed. 

The circuit shows several by-pass tubes in parallel. There are 
several reasons for this. The typical triode used in this circuit has a 
plate resistance of 10,000 ohms, 15 volts on the grid at 135 volts plate. 
That means that only a 15- volt d-c swing is required from cut off to 
minimum resistance. This can readily be produced across resistor 
R and filtered with condenser C and a very short time constant can 
be obtained. 

On the other hand, if a tube is used such as one with 800 ohms' plate 
resistance, 60 volts on the grid and 250 on the plate, the d-c swing 
must be (>0 volts. This would entail, assuming that enough current 

Jan., 1940] 



would flow from the diodes, a larger resistance and larger condenser. 
The time-constant would be very large and the action would be very 
sluggish. In addition, the grid would need about 100 volts negative 
to block. This means larger batteries. 

Therefore the multiple-tube arrangement was evolved. By in- 
creasing the number of tubes used the resistance of the by-pass could 
be made any minimum value. This gives a wide latitude of resistance 
change and works out very nicely. 

Feedback. The by-pass line with its tubes is actually a method 
for introducing to the main signal line the rectified d-c "hash." 

SOO 1000 


FIG. 7. Curves of selective by-pass system. 

Due to a fortunate choice of constants for the various elements, the 
by-pass line, in the amplifiers now in use, is coupled directly as is 
shown in Fig. 7. With this arrangement the feedback is so small that 
no distortion is detectable with the ear. 

Before this combination was arrived at several feedback suppres- 
sion devices were developed. 

These may be used if too much distortion due to feedback occurs. 

Reflection. Reflection at the by-pass line in the system, due to 
the major changes in the impedance of the by-pass line, does not 
occur in detectable form until extremely high signals occur. 

Full-Wave Rectification, A word should be said as to why full- 
wave rectification is used in place of the mechanically simpler half- 

72 S. L. REICHES [J. S. M. P. E. 

wave. This is because less filtering is needed in full-wave rectifica- 
tion, and second, it was felt that some advantage would be gained 
if the envelope were a more exact replica of the audio than could be 
obtained with half -wave rectification. This is because of the non- 
symmetry of the audio about the reference axis. 

Range of System. This arrangement is capable of by-passing as 
much as 30 db at 10,000 cycles. 

Microphonts. Due to the effect of the amplifier it seems that 
best results are obtained when a "live" microphone, such as the 
"eightball," is used. 

Other microphones can be used but require some adjustments in 
the response of the line amplifiers. These adjustments consist in 

POC-AMP. PowtR Ant 

FIG. 8. Typical amplifier arrangement for radio and 

raising the high-frequency response of the line amplifier to compen- 
sate for the bass response of the microphones. 


Although some measure of success was had in the design of the 
required amplifiers, the problem remained to develop a technic for 
the use of the amplifiers. 

In the usual recording set-up, whether for disk records, radio, or 
sound-film, the microphones are judicially placed according to past 
experience. The microphones are fed to pre-amplifiers. The output 
of the pre-amplifiers is fed to the line amplifiers and then to the re- 
cording unit. Gaining is usually done before the line amplifiers. 
The gain is adjusted to the requisite level and is independent of the 


input sound energy. This can not be done with the automatic 
response control (abbreviated A. R. C-) amplifier. 

General Consideration. In this amplifier the frequency character- 
istic is a function of the voltage applied to the grids of the tubes. 
Because of this the input to the device must have such a value, for 
a given input to the microphone, that the proper response curve will 
appear for the A. R. C. amplifier. Also, if it is desired to have the 
amplifier correct for the distortion produced by the system limit, 
all gaining must be done after the A. R. C. amplifier. 

Typical Radio Set-Up. In radio there are two types of sound to 
be considered, speech and music. Investigation will show that the 
error produced in radio for speech occurs because of a higher reproduc- 
tion level than incident level. Music, as was pointed out before, 
produces errors in the other direction. 

FIG. 9. Typical arrangement of amplifiers for sound 

It was decided that the error for speech was not worth correcting 
because the error is so small in most cases, and also, in a musical 
program, the announcer was only a small part of the show. There- 
fore the arrangement of Fig. 8 was used. 

As is typical, the announcer feeds through a separate microphone 
and amplifier system to the main control board. The musical pro- 
gram is fed through a preamplifier to the A. R. C. then to a power 
amplifier and finally to the master control panel. This arrangement 
proves satisfactory for the studio and remote pick-up. 

Network Pick- Up. Here another problem presents itself because 
the announcer's voice is fed to the chain at the same level as is the 
music. Therefore, the A. R. C. will respond to the voice in the same 
way as music does. This causes bad distortion for voice. The only 
way in which this type of pick-up can be used is to by-pass the A. R. C. 
whenever the announcer talks. 

Records. The same arrangement is used for records with the ex- 
ception of the announcer's microphone. 

74 S. L. REICHES [J. S. M. P. E. 

Sound-Film. In sound-film recording the same general considera- 
tions presented for radio hold except for the announcer's microphone. 
Here, however, the amplifier line-up is quite different. As was shown, 
the type of correction is different and a different A. R. C. is needed. 
A recommended installation is shown in Fig. 9. 

Determination of Levels. To determine the average levels about 
which correction would be made, advantage was taken of the ampli- 
fier as a loudness meter at the recording end, and a typical amplifier 
loudness meter was used at the reproduction end. The automatic 
response control amplifier was calibrated from the commercial loud- 
ness meter. Loudness was measured at the distance of the micro- 
phone from the source. 

In reproduction, the distance from the reproducer at which loud- 
ness is measured is debatable. In a receiver a distance of six feet 
was used. For sound-film, possibly, the center of the auditorium 
in line with the screen is a good point. 

The desirable method would be to use a constant-energy sound 
which, when placed at the desired distance from the microphone and 
adjusted to the proper loudness, would create linear response curves 
for the A. R. C. This linear condition is easily determined by the 
magnitude of the by-pass triode plate current. 

Re-Recording. In re-recording existing sound copies (either disk 
records or film) the improvement can be only within the volume 
range of the present record. The problem here is to determine in- 
put levels from the pick-up to the A. R. C. At present disk records 
are copied with the by-pass plate current meter deflecting an amount 
determined by the type of sound being re-recorded. 


It is believed that this paper presents some new concepts on the 
problem of recording sound whether in radio, for wax records, or 
sound-film. The theory discussed is naturally quite general because 
of the many different ear-sensitivity curves. Yet it is sufficiently 
accurate so that an amplifier with only approximate correction curves 
records sound that, in the opinion of the great majority of listeners, 
is more realistic than that provided by present-day methods. It has 
been found that corrections are worth more, the greater the difference 
between the incident level and the reproduced level. 

The greatest handicap in this work was the lack of live pick-ups 
that could be used for test purposes. Because of this it is not pos- 


sible to say that the A. R. C. improves everything that passes through 
it. Much more work must be done along these lines. 

The shape of the response curves has been shown to be moderately 
critical. It has been found that the closer these curves approach the 
ideal the greater is the apparent improvement. Much more work 
can be done to improve these curves. No mention has been made 
of public address systems and electrical musical instruments. Some 
work has been done at the level of public address systems in conjunc- 
tion with sound-film. However, a complete analysis can not be 
presented as yet. 

Acknowledgment. The writer expresses his appreciation to 
Professor John Martin, of the Department of Electrical Engineering, 
Case School of Applied Science, for his many suggestions; also to 
Mr. R. Morris Pierce, Chief Engineer, and Mr. Lawrence Shipley, 
Assistant Chief Engineer, respectively, of Station WGAR, for their 
cooperation in permitting the use of the station's facilities and for 
their encouragement and direct solutions of many of the problems 
that have turned up in this work. 


Summary. In all motion picture photography and projection, lenses of high 
relative aperture must be used. However, on account of the small size of the amateur 
frame, the focal length is short, and the linear aperture of the lens is therefore small, 
resulting in considerable depth of field. Thus in cine work, great lens speed is not 
automatically associated with small depth, as is the case in ordinary photography. 

Moreover, as the entire motion picture frame must be seen by the eye at a glance, 
the angular field covered must be much smaller than in still pictures which may be 
examined critically and deliberately. This fact is of the greatest assistance to the 
lens designer because high aperture and field are inevitably somewhat incompatible, 
and types of lens construction which favor aperture generally cover a relatively small 

Perspective considerations usually require a projection lens covering only about 
half the angular field covered by the taking lens, which fact enables projection lenses 
of very high relative aperture to be made. Some of the types of construction com- 
monly used in amateur cine lenses are described. 

Perspective Considerations. The lenses used in amateur cine cameras 
imitate those used in professional work with 35-mm film in one im- 
portant respect, namely, angular field coverage. If a professional 
lens of 50-mm focus covers a 35-mm sound-frame measuring 16 X 22 
mm, having a diagonal of 27.2 mm, the angular semifield to the corners 
of the picture is 15.2 degrees. In the amateur sizes, with a 16-mm 
frame having dimensions of 7.5 X 10 mm, the picture diagonal is 
12.5 mm and the semifield to be covered by lenses of various focal 
lengths is as indicated in Table I. The cine-eight frame is just one 
quarter of the 16-mm frame, and has a diagonal of 6.25 mm. 

It is thus clear that lenses of those focal lengths which are normally 
used in cine work operate at a much smaller angular field than the 
lenses on ordinary still cameras, which usually cover about 24 degrees. 

The reason for this is that the eye can take hi a semifield of only 
some 12 to 14 degrees at a glance. Thus in a motion picture in 

* Presented at the 1939 Fall Meeting at New York, N. Y. ; received October 
13, 1939. 

** Eastman Kodak Co., Rochester, N. Y. 



which the eye must follow action covering possibly the entire frame, 
a semifield of only 12 to 14 degrees is desirable at the observer's eye 
during projection. Now, the rules of correct perspective rendering 
tell us that all angular relationships in the final picture as seen by the 
observer should be the same as those which existed in the original 
subject as seen from the camera lens. Thus if a scene appeared 24 
degrees wide to the original camera, its projected image should be 
viewed from such a distance that it also appears 24 degrees wide to 
the audience. 

Since the average position of the audience may be perhaps midway 
between projector and screen, it is necessary that the focal length of 


Angular Semifield Coverage (Degrees') in Amateur Cine Cameras 
Focal Length 16 Mm 8 Mm 


26.2 13.8 

22.6 11.8 

17.4 8.9 

14.0 7.1 

9.3 4.7 

7.1 3.5 


the projector lens should be about twice that of the taking lens (Fig. 1) 
and the angular field covered by it is therefore only about half that 
covered by the taking lens. Thus if a focal length of 25 mm is 
adopted as the standard for taking lenses, the standard home projection 
lens should be of 50-mm focus. Of course in a much longer hall, if 
the audience is fairly close to the screen, longer focus projection lenses 
should be used. 

The wider (25 degrees) field of a still camera is justified because, 
the picture being still, the eye has an opportunity to scan it and study 
each part separately. But in this case also, the perspective will be 
correct only if the eye is at the correct viewing distance, which is 




























[J. S. M. P. E. 

such that the angular size of the picture appears equal to the angular 
subtense of the original subject at the camera lens. This distance is 
equal to the camera lens focal length multiplied by the enlargement 
ratio in the print. 

Choice of Lens Aperture. In all photographic lenses, a high relative 
aperture is desirable to shorten exposures. In cine work, however, 
the exposure is ordinarily fixed, and then a high aperture is valuable 
to facilitate photography in a poor light or indoors. Fortunately, 
the use of short focal lengths enormously increases the depth of field, 





FIG. 1. Perspective of Projected Images. 

because depth of field depends only on the distance of the object 
from the camera and on the linear aperture of the lens. Thus an 
//2 lens of 25 mm focus has a clear linear aperture of J 2.5 mm, and 
this lens therefore has the same depth of field as a 100-mm lens at 
//8. In cine-eight cameras, the standard focal length is only 12.5 
mm, and an aperture of //2 has the same depth of field as a 100-mm 
lens at //16. 

Sharpness of Definition. If the audience is situated at approxi- 
mately the correct center of perspective for the projected picture, 

Jan., 1940] 



the angle relationships on the projection screen as seen by the viewer 
must be the same as the angle relationships in the original film as seen 
from the second nodal point of the taking lens. 

Suppose therefore we are at the limiting stage of critical definition 
when the screen image is just beginning to lose sharpness as seen by 
the audience. Then each point will be imaged on the screen as a 
tiny blur-circle subtending an angle of about 1 in 2000 (1.7 minutes 
of arc) at the eye. This corresponds to a circle of confusion on the 
film of diameter equal to Vaooo of the focal length of the taking lens, 
and if that lens has a focal length of one inch (25 mm), the circle of 
confusion will be just on the verge of spoiling the definition when it 
has a diameter of Vzooo inch or Vso mm. 

The graininess of reversal films is of this 
order; hence graininess will usually be just 
visible on the screen when properly viewed, 
and it will certainly be visible to a member 
of the audience who is too close to the screen. 
However, as the grain pattern in successive 
frames is entirely different, persistence of 
vision tends to smooth it out and the graini- 
ness is therefore usually not objectionable. 
Moreover, a moving object may be imaged 
decidedly unsharply as the motion, together 
with persistence of vision, helps our inter- 
pretation to a great extent and makes very 
sharp definition unnecessary. 

We may therefore conclude that a taking 
lens must form an image-point which is less 
than Vso mm diameter if satisfactory definition is to be obtained. 
This applies, of course, over the entire frame, but it is less impor- 
tant at the corners since the major attention is generally paid to 
details and action in the center of the picture. 

Since sharpness of definition is fundamentally an angular quantity, 
it follows that if a particular lens formula gives acceptable definition 
in one size it will also be acceptable in all sizes, provided the final 
picture is viewed from the correct perspective center. Thus cine 
lenses gain nothing on account of their short focal length, for an 8- 
inch lens must give a sharpness corresponding to a circle of con- 
fusion of about Vio mm, and a 1-inch lens must give Vso mm. The 
same lens formula made in 8-inch and 1-inch sizes would therefore 

FIG. 2a. Triplet lens. 
FIG. 2b. Tessar lens. 



[J. S. M. P. E. 

provide lenses which are equally good, if the final viewpoint is always 
correctly chosen. 

However, in using interchangeable cine-camera lenses, say from 
1 inch to 6 niches in focus, the pictures are viewed in succession from 
the same viewpoint, and thus the 6-inch lens and the 1-inch lens must 
both give a Yso-mm circle of confusion. This puts a great demand 
upon a 6-inch lens, as it must be relatively 6 times as good as the 
1-inch lens. On the other hand, the angular field is only one-sixth, 



FIG. 3a. Petzval type//1.6. 
FIG. 3b. Petzval type//1.3. 
FIG. Zc. Petzval lens with field flattener. 

so a design may be employed for these very long-focus lenses which 
stresses central definition entirely at the expense of field. 

Types of Lens Construction. Any type of lens may be used for 
cine purposes, but as high relative apertures are desirable, the actual 
choice of lens types is really rather limited. However, since the 
normal angular field is small (14 degrees), a number of types may be 
used which could never be made to cover the 24 degrees demanded in 
a still camera. It is well known that aperture and field are more or 
less incompatible, and we may secure much higher apertures if we 
are willing to sacrifice angular field. 

Jan., 1940] 













For apertures of //3.5 or//2.7, lenses of the regular triplet or Tessar 
types may be used (Figs. 2a and 26). In these cases, a field as wide 
as 22 degrees may be covered without difficulty, and such forms lend 
themselves to the construction of "wide-angle" cine lenses which are 
lenses covering a wider semifield than 14 degrees. 

Above f/2.7, the choice of lens type becomes much more limited. 
The Petzval portrait lens (Fig. 3o) is an obvious choice and has been 
modified to give apertures as high as //1. 3 (Fig. 36) but this really 
does not cover as much as 14 degrees. Indeed, at apertures higher 
than //2, this type will not give acceptable definition much beyond 7 
degrees from the axis. The angular field 
may be increased by the use of a con- 
cave "field flattener" lens placed close 
to the image plane (Fig. 3c). 

In the normal type portrait lens with 
cemented components, all the correction 
is done by the two negative cemented 
surfaces. Thus the crowns must have a 
lower index than the flints, borosilicate 
crown and ordinary dense flint being 
the usual choice. However, if the two 
doublets are separated, and the crowns 
and flints are "bent" independently, a 
strong negative air-lens may be inserted 
between the crowns and flints, and a 
high-index barium crown may then be 
employed. This at once reduces the 
Petzval sum of the system and enables a much wider field to be 
reached before the astigmatism begins to spoil the definition. 

The Cine Kodak //1. 9 lens is of this type (Fig. 4a) and was first 
described by Frederick. 1 By extending this process, the R-Biotar 
//0.85 was designed by Merte" in 1934 2 (Fig. 46). 

The Petzval type can be used as a taking lens of high aperture in 
the longer focal lengths, provided a semifield of less than 7 or 8 de- 
grees must be covered. Such lenses give brilliant pictures, free from 
haze and scattered light, and have been extensively used in focal 
lengths of 2 1 / 2 inches or more on 16-mm film. 

If an aperture greater than about f/2 is required, at a field of 14 
degrees some other type of construction must be adopted. Innumer- 
able attempts have been made to solve this problem, and almost 

FIG. 4a. Kodak //1. 9 cine 

FIG. 4b. TheR-Biotar//0.85. 



[J. S. M. P. E. 

every lens manufacturer has made cine lenses having apertures any- 
where from //0.85 downward. A popular type is the Planar de- 
scribed by Rudolph 3 in 1897 (Fig. 5a). 
The original Planar worked at //4 and 
covered a 25-degree field. It was sym- 
metrical, and used glasses such as were 
available at that time. In 1921, Lee 
in his "Opic" lens 4 raised the aperture 
to f/2 by departing from strict sym- 
metry and using other types of glass. 
In 1927, Merte* developed the //1. 4 
"Biotar" 6 (Fig. 56) by introducing 
stronger surfaces and still greater asym- 
metry, to cover a field of 14 degrees 
for cine purposes. Many other modi- 
fications of this type have been made 
with varying degrees of success. This 
type of construction leads ordinarily 
to a very flat field with good axial 
spherical correction, but it tends to 
suffer from considerable spherical aber- 
ration in the oblique pencils. This 
does not detract from the definition, 
but it may introduce some slight haze 
and lack of contrast in the outer parts 
of the field, especially on over-exposure. By using two single lenses 
in the rear component, some further benefits may be derived, as 
in the Xenon //1. 3 lens (Fig. 5c). 

FIG. 5a. 
FIG. 5b. 
FIG. 5c. 

Planar //4. 
Biotar //1.4. 
Xenon //1.3. 


FIG. 6. A telephoto lens. 

Telephoto and Wide-Angle Lenses. The true telephoto lens com- 
prises two widely separated members, the front being positive and 

Jan., 1940] 



the rear negative. The "telephoto magnification" is then often 
defined as the ratio of the overall focal length to that of the front 
member alone. True telephoto lenses have been made at apertures 
approaching //3.5, with a telephoto magnification usually of about 
2.0. The advantage of the telephoto lens is of course its compact- 
ness, but the disadvantages are that with this type of construction 
it is not possible to secure either a high aperture or the very finest 

FIG. 7. Telephoto and wide angle attachments. 

degree of aberration correction. Consequently, in the short focal 
lengths required in amateur cine work, it is common to use a normal 
lens in place of one of the strictly telephoto type. 

A "wide-angle" lens in cine work implies one covering a semifield 
of more than 14 degrees. Thus a regular camera lens covering 24 
degrees may be made in 15-mm or 20- mm focal length as a "wide- 
angle" lens for 16-rnm film. 

)l." Telephoto and Wide-Angle Attachments. These are merely specially 
designed forms of Galilean telescope intended to be added in front of 

84 R. KlNGSLAKE [J. S. M. P. E. 

the regular lens on the camera. If the positive lens is in front, a 
magnifying action is produced, and if the negative lens is in front, the 
result is a wide-angle effect. 

In some of these attachments, focusing may be done by increasing 
the separation between the components. Of course, when such an 
attachment is in use, the regular focusing scale on the camera lens 
becomes meaningless. 

It should be noted that these attachments do not affect the /- 
number of the lens, but only its equivalent focal length and hence its 
angular field. 

Kodak 16- Mm and 8- Mm lenses. In the 16-mm range is the follow- 
ing series of taking and projection lenses. (After each lens is indi- 
cated the number of the illustration in which the type of construction 
is indicated.) 

Taking Projection 

Focus Aperture Fijr. Focus Aperture Fig. 

15 mm f/2.7 2a 1 inch //1. 9 4o 

25 mm 1.9 4a iVz 2.0 4a 

2 inch 3.5 2a 2 1.6 3o & 3c 
2'A 2.7 3c 3 2.0 3a 

3 4.5 6 4 2.5 3a 

4 2.7 3o 4 1.6 3a 
4V 4.5 6 

6 4.5 6 

For the cine-eight, the following lenses are available : 
12.7mm //3.5 2a 1 inch //1. 6 3a 

13 2.7 2o 1 2.0 3a 

13 1.9 4c 1 ' 2.5 3a 

1'A inch 4.5 6 

> U. S. Pat. 1,620,337. 
1 U. S. Pat. 1,967,836. 
U. S. Pat. 583,336. 
4 Brit. Pat. 157,040. 
U. S. Pat. 1,786,916. 


MR. TOWNSLEY: You stated that probably the reason for wanting a narrow 
angle for motion picture work was perspective. For 8-mm and 16-mm equip- 
ment particularly, is it not probably true that one of the major considerations is 
the necessity for getting the shutter and aperture plate between the back of the 
lens and the film, for normal construction? 

MR. KINGSLAKE : The back focus must also be considered, but I imagine that 


this field-angle was determined before the 16-mm Cine camera came into the pic- 
ture at all. The small angle was used in the early days when back focus was not a 
problem and the 8- and 16-mm designers desired to retain that small field. If a 
wider field-angle were demanded, high speed would become difficult, which fact 
alone would force us to use the smaller field, even if other considerations had not 
come in first. 

MR. MITCHELL: In the early days of 16-mm film one of the principal points 
made when the one-inch lens was used, which was standard equipment, was that 
the lens would give the equivalent perspective of the two-inch lens used for the 
standard 35-mm film in the Hollywood studios. 

MR. KELLOGG : Referring to the device consisting of a positive and a negative 
lens, which can be put in front of the camera to enlarge the field or give a wide- 
angle effect, it appeared in the diagram you showed that if parallel light enters 
either lens it issues from the other as parallel light. When this condition is ful- 
filled, is it not true that the system is reversible provided both lenses are of ade- 
quate diameter, thus serving as either a wide-angle attachment or as a telephoto 

MR. KINGSLAKE: You can reverse them, but mechanical considerations are 
the usual limitation. Some lenses are mounted a long way down the cell, and it 
is necessary to push the back lens of the attachment close to the camera lens, so 
that you would need specially mounted camera lenses to accommodate a reversible 
attachment. Moreover, the corrections which would have to be included would 
make an interchangeable attachment expensive, and it would be cheaper to have 
two lenses than one, although it would of course be more convenient to have one. 
In ordinary cameras there is enough room and you could turn the lens around from 
front to back quite easily. 

MR. ROGER: Have experiments been made with regard to the optimum in 
sharpness, first, with the mechanism still, and, second, with the mechanism operat- 
ing normally? Since in a motion picture camera, with its gears and intermittent 
motion, a slight vibration may be felt in one's hand, this vibration might perhaps 
be transmitted to the lens and so cause lack of sharpness on the film, however slight. 

MR. KINGSLAKE: That is true, but ordinarily Cine lenses are manufactured 
and treated with exactly the same procedure as for ordinary lenses. They are 
tested first on a lens-testing bench, and they are tested also for motion pictures, 
and comparisons made very carefully. The lens taken by itself without any film 
at all should give as good an image as a camera lens gives. 

MR. MAURER: One of the defects I have occasionally encountered in 16-mm 
photography is the tendency of the reflections from the internal surfaces of a lens 
to form a round image in the center of the picture, possibly an image of the front 
of the lens hood, which makes a bright flare spot in the picture. This occurs with 
a number of widely used lens types. Can you tell us whether modern lens de- 
signers have been making any attempt to expand the utility of the type of lens 
that is represented by the well known Dagor and Protar types, in which the com- 
ponents of the two sections of the lens are all cemented together, thus cutting 
down the number of reflecting surfaces. Most lenses of those types have been 
pretty slow, but if I am not mistaken there was one made in England that was 
pushed up to an aperture off/4. It would seem as though, with further develop- 
ment, that type of lens might become extremely useful. 

86 R. KlNGSLAKE [J. S. M. P. E. 

MR. KINGSLAKE: That is true. You can make those types up to //4. In the 
original Dagor design the contact surface inside has almost reached the hemisphere 
at//4, and the only way you can improve it would be to use high index flints and 
low index crowns, but the result would not be very satisfactory. 

The residual zonal aberrations are very large in that type, whereas they are 
very small in the Petzval type, which is why that type has been used for the higher 
apertures. The question of a flare spot is usually attributed to an image of the 
iris formed by reflexes in the back member of the lens. The flare spot appears 
only at very small stops //16or//22 and in many types it seems to be practically 
unavoidable. On the other hand, other types of lenses do not have a flare spot. 
A notable case is the Sonnar, which has recently been introduced by Zeiss. The 
back part of that is a cemented triplet, and it behaves like the Dagor type in not 
forming a flare spot. The problem is a very live one to lens designers, but it 
seems extremely hard to avoid it, although by chance we may land on a lens which 
is good otherwise and does not have a flare spot. Ordinarily the flare spot is con- 
sidered of secondary importance. It can be avoided by not taking a photograph 
of a dark subject against a bright background. If you take a photograph of a 
person in the middle of a picture, in front of a window, for example, the flare spot 
may be quite marked; but if you take care that there is not a bright glare of light 
on the outside of the picture and a dark center, the flare spot ordinarily is not 

MR. BRADY: Assuming that we can eliminate the intermittent movement, 
what advantage would that be to you as a lensmaker? 

MR. KINGSLAKE : I do not believe it would make any difference. We merely 
consider the lens as covering the full field of the picture. 

MR. BRADY : Would eliminating a rear or front shutter help ? 

MR. KINGSLAKB: Sometimes not having a rear shutter would have a slight 
advantage. That would, however, not be ordinarily important except in Cine 8 
cameras where the back focus is getting very small. 

MR. FAMULENER: Has work been done in applying to lenses some of the non- 
reflecting wax coatings that have recently been worked out? 

MR. KINGSLAKE : That has been tried a good many times, but there is always 
the question of getting at the surface. I imagine you could put the wax layer on 
the inside of a lens with great ease, but how long it would stay there, I do not know. 
One would not like to sell lenses with such an unstable coating on them as that. 

MR. WALKER: What is the difference between a lens which is corrected for 
color and one which is not? 

MR. KINGSLAKE: All lenses today are corrected for color by the use of more 
than one kind of glass in the lens. Achromatism was introduced 150 years ago. 
You could not sell a lens today in which the color was unconnected. 

MR. FLORY: It has been my experience that in both taking and projecting 
16-mm motion pictures, the quality of the taking lens is far superior to that of the 
projector lens. Is that true? 

MR. KINGSLAKB : In what respect? 

MR. FLORY: Flatness of field on the screen, for example. Sixteen-mm pro- 
jector lenses are made to sell for a low price, and I have not been able to buy a 
lens which satisfies me. The cost of most of the projector lenses runs from $20 to 
$25, and I think some work should be done on manufacturing projector lenses 


that are more comparable, so far as quality goes, to the camera lenses. We will 
spend $100 for a camera lens, and then buy a projector lens which is relatively 
inexpensive, as a result of which, it seems to me, we lose a great deal of the quality 
in projecting after we have gone to a great deal of trouble in producing a good 
negative and print. 

MR. KINGSLAKE: That is true. Most projection lenses are of the simple type 
which has a field of only about 5 degrees. The solution is to put on a field flat- 
tener, and many of the better projectors are now equipped with such field flat- 
teners. I agree that it seems poor economy to have a cheap lens on a good pro- 
jector. Probably in time to come field flatteners will be supplied on all projectors 
as a matter of course; indeed, I imagine the public will demand it. 

MR. PALMER: Can we assume that with a lens that is corrected for color the 
visual focus and chemical focus are both in the same plane? 

MR. KINGSLAKE: That is usually an entirely safe assumption. Sometimes 
lenses get by through carelessness in design or accidents in manufacture, in which 
there is a slight difference between visual and photographic focus, but the aim is 
always to eliminate that small difference. On the other hand, it seems impossible 
to bring all the colors to a common focus. The best you can do is to bend the 
spectrum around so that the blue and yellow coincide, and let the other colors 
take care of themselves. Ordinarily, that is quite a safe compromise. Some- 
times the blue is bent around too far and falls beyond the red, and then you get a 
little chemical focus; or it may be a little under corrected, in which case the blue 
coincides with the green, and the red is too long. If you do find a difference, it 
is generally either accident in manufacture or carelessness in design. 

MR. FRITTS: In judging quality it is well to note the different conditions under 
which cameras and projectors are operated. I believe Mr. Kingslake's paper 
presupposes a flat film in each case, but with increase in light in the projection 
beam it becomes increasingly difficult to hold the film flat. Perhaps the difference 
between projection quality and camera quality might hinge upon that point, 
rather than upon any consideration of the lens itself. 

MR. ROGER: In still photography, one uses special apochromatic lenses, es- 
pecially in the graphic fields, for color reproduction. They use apochromatic 
lenses for getting the best color results. Are apochromatic lenses also available 
for motion pictures and 16-mm motion picture photography? 

MR. KINGSLAKE: The name "apochromatic" is very unfortunate. It was 
originally applied to microscope objectives, in which an attempt was made to 
bring three colors to a common focus. In motion pictures, some lenses are avail- 
able in which the residual color differences have been considerably reduced, and 
such may sometimes have been called "apochromatic." The so-called apochro- 
matic process lens is surprisingly well corrected in this respect, but more espe- 
cially in respect to the equality of picture size in all colors. The name "apochro- 
matic" has often been used merely as a sales point. 


Summary. Proposals have been received from the ISA Secretariat for Interna- 
tional Standardization of raw-film cores; 16-mm sound-film; projection reels; pro- 
jection reel boxes; 8-mmfilm dimensions; and definition and marking of safety film. 

Most of these proposals differ from the SMPE standards only in tolerance. Some 
of the tolerances appear to be unimportant and some important. The European 
practice for projection reels differs so widely from the American practice that it is 
deemed impossible to come to an international agreement. Standardization of 
16-mm projection reel boxes appears to be outside the range of useful standardization. 

The international standard definition of safety film has been cleared up in all 
points except the question of nitrogen content. 

The question of sound-track dimensions for 35-mm and 16-mm film was clarified, 
to a considerable extent, at the Hollywood meeting of the Committee but no definite 
conclusions have yet been reached. 

No satisfactory standard for 16-mm sound-film sprockets has yet been attained. 

The work of the Standards Committee, since the last meeting, 
may be divided into four classes : 


An attempt has been made to come to an agreement with the 
German Secretariat of the International Standards Association and 
to bring their proposed standards and the American standards hi line. 
In nearly all cases the differences are minor ones involving tolerances. 
Our last reply to their proposal was sent shortly before the war began, 
and it appears likely that further developments in the industry will 
modify the standards or tolerances before there can be any hope of 
an international agreement in these matters. 


We have prepared our standards for submission to the American 
Standards Association and adoption by them as American standards. 
In most cases, this has meant solely a change in form rather than 
any change in the actual standards themselves. The American 

* Presented at the 1939 Fall Meeting at New York, N. Y.; received October 
12, 1039. 



Standards Association objects to any reference in their publication 
to the organization sponsoring these standards, and this has resulted 
in considerable editing of our standards drawings. 


(a) Sound-Track Dimensions. The most important revision of 
old standards is the work on sound-track dimensions being carried on 
largely by the Academy of Motion Picture Arts and Sciences. It 
will be remembered that about a year and a half ago a proposal was 
made to increase the width of the variable-area sound-track at 100 
per cent modulation from 0.071 to 0.076 inch. There were serious ob- 
jections to this proposal, and a Committee of the Academy has been 
working for some time, attempting to reconcile the various difficulties 
with one another. One proposal, which at present seems to take 
care of many of the difficulties, is to use the width of 0.071 inch for 100 
per cent modulation, and to specify a maximum width of 0.076 inch 
for the sound-track. This maximum width would be fixed by what- 
ever mechanical features are involved in the sound equipment, so 
that, in cases of overrunning the 100 per cent width, no distortion would 
occur up to 0.076 inch provided no further errors were introduced 
in the printing or projection. This practice appears to be the actual 
practice in use at the present time in the majority of cases. The 
question of the separation between the two halves of the push-pull 
sound-track has not yet been satisfactorily determined, although 
the prevailing practice still appears to be to allow 0.006 inch. 

The standardization of 16-mm sound-track dimensions should be 
based on those for 35-mm, and this revision is also awaiting the work 
on 35-mm sound-film standards. 

Definition of Safety Film. The question of a standard definition 
of safety film has, in the past, been very much of an international 
question, and probably no satisfactory agreement will be obtained 
until the end of the war. Although we are recommending to the 
American Standards Association our old standard method of deter- 
mining burning time, we are in practical agreement with the European 
countries in regard to a new method of determining burning time 
and defining safety film, except that we are still not in agreement as to 
whether or not a limit should be placed upon the cellulose nitrate con- 
tent of safety film. The international recommendation is that this 
limit should be placed at a figure corresponding to 0.36 per cent nitro- 
gen, whereas some in this country maintain that no specification 









25.90 * 0.20 
50.00 0.25 
15.50 =t 0.50 

1.020 0.008 
1.968 0.010 
0.610 0.020 

Recommended Practice 


16.70 0.30 
4.00 0.20 

0.657 * 0.012 
0.157 =* 0.008 

Bore A to fit freely to Hub 25.40 =*= 0.1 mm or 1.000 * 0.004 inch diameter. 

should be drawn on the analysis of the film base, but should be drawn 
solely on its performance. 


16-Mm Sound Sprockets. The question of standards for 16-mm 
sound sprockets is still under consideration. Some of the manu- 
facturers of projectors maintain that the design of the sound sprockets 
depends so much on the design of the rest of the projector that there 
is very little that can be usefully standardized. 

Reduction Ratio for 16-Mm Reduction Prints. This also depends 
somewhat on the 35-mm sound standards and is awaiting that 


Future Work. At the last meeting of the Standards Committee 
several items were brought up that appeared to warrant future inves- 
tigation. These are: 

(1) A standard method of rating loud speakers. 

(2) A standard method of measuring flutter. 
(5) A standard method of rating amplifiers. 

(4) A standard method of measuring auditorium acoustics. 

(5) A revision and elaboration of our glossary of technical terms. 

(6) A standard method of measuring screen brightness. 

After initial approval by the Committee, a letter ballot was recently 
taken of the entire Standards Committee on the Projection Room Plans 
prepared by the SMPE Projection Practice Committee, and the raw- 
stock core specifications prepared by the SMPE Standards Committee. 
The Projection Room Plans were published in the JOURNAL of the 
Society, November, 1938. The other project is reproduced herewith. 

Having received unanimous approval of the two projects by the 
letter-ballot of the Standards Committee, publication of the fact is 
made hereby. If within sixty days of publication of this issue of the 
JOURNAL, no objections to these proposals arise from the industry, the 
proposals will be transmitted to the Board of Governors for valida- 
tion as SMPE Standards, after which they will be transmitted to 
the Sectional Committee on Motion Pictures of the American Stand- 
ards Association. 

E. K. CARVER, Chairman 












MR. CRABTREE : With regard to the glossary, why not take a vote and let the 
majority decide whether they want to include all terms, or restrict the glossary 
to the "highbrow" terms? The Nomenclature Committee could split up its re- 
port in two parts, dealing with the "slang" terms in an appendix. A glossary is 
useless unless it describes what a "wow" is, for example. 


MR. OFFENHAUSER: In the committee meetings that question was discussed, 
and a term about which much discussion arose was the word "jeep." A "jeep" is 
well understood by television engineers, but has not yet been adopted by motion 
picture engineers. Should we include the glossary such a term as "jeep" ? 

MR. CRABTREE: I would hold that term in abeyance for a while; but there are 
many other terms that have come into common use and are not in our glossary. 

MR. Ross: The non-technical expressions might be added to the definitions of 
the technical terms. For example, the word "bloop" should be included as a non- 
technical term, and the laboratory equivalent stated also. 

MR. WILLIFORD: My feeling is much the same as Mr. Crabtree's. When we 
do not know what people are talking about, we go to a glossary, expecting to find 
there the things we do not know. Whether we like the use of slang in our in- 
dustry or not, it is here; and when the people who use a thing give it a name, that 
is the name the thing is going to be known by. I could give you one or two classic 
examples of companies trying to get the public to stop calling things by names the 
companies did not want them known by, and spending a lot of money to get new 
names accepted, but it could not be done. So if it is a "jeep" in the television 
industry, it is probably going to be a "jeep" in ours. The discussion we are hav- 
ing here ought to be of some help to the Committee in determining their policy, 
but I do not believe, as a point of order, we ought to vote on it, although everyone 
who has an opinion on it ought to express it. 

MR. KELLOGG: It would be helpful if we could standardize our definition of 
"wow" or "flutter" not the word but the thing itself. Measurements of speed 
variation may be metered in either of two ways. It is very common to use a 
meter which will measure the rms deviation from an average speed. It is also 
very valuable, for finding out what is the matter with a machine, to make an 
oscillogram showing from instant to instant what the speed has been. That will 
show whether it has been consistently above or below the normal speed, as well as 
the fluctuations. We have used a "wow-meter" of the oscillograph type for most 
of our studies in Camden, but on the West Coast it has been the general practice to 
use the rms type of flutter-meter. Some confusion is likely to result in reporting 
the performance of the various machines tested. The oscillogram does not give a 
ready indication of rms deviations, but has been used for showing the overall 
change from minimum to maximum speed, in a period of about six seconds. 
Clarifying definitions of wow and flutter would be a valuable service. I suggest 
that the term "rms flutter" be employed to designate the flutter as measured on 
an rms type of meter, while "overall wow" or "overall flutter" be used to designate 
the extreme fluctuations. The overall wow is usually from four to six times as 
large as the rms flutter. Even if the speed changes were a pure sine wave, the 
overall value would be 2.8 times as large as the rms value. 

MR. IVES: If the Committee finds it impracticable to attach its authority to 
the definitions within a reasonable time for members to make use of such a new 
edition of the glossary, might not the term be published as being under the con- 
sideration of the Committee, with such connotations as have been suggested by 
responsible persons familiar with the particular branch of the field wherein the 
terms are used frequently? The terms will then be available in the JOURNAL, 
even if they do not have the backing or full authority of the Committee. 

MR. OFFENHAUSER: The difficulty with a procedure of that sort is that th 


average reader will glance at the first few terms of the glossary and then start 
looking at the next article. It will be difficult to get each of the members who 
would be concerned particularly with the glossary to take an interest in it at the 
time it is printed. 

MR. MORGAN: Perhaps a questionnaire would be more effective than using the 

DR. CARVER: That is what we had intended to do to send parts of the glos- 
sary to those who were especially interested, with a questionnaire. 

MR. DAVEE : It seems to me that a word like "jeep" is a little far-fetched. The 
word "wow" conveys some idea to us; but the word "jeep," as I understand it, 
means a complete television transmitter and receiver in practically one unit. 
Now, if "jeep" conveys anything as far as that particular adaptation is concerned, 
I can not see it. Such words should be left out of our glossary, and maybe a word 
better adapted to the meaning should be substituted. We should not let the in- 
dustry go "haywire" on these terms. 

MR. Ross : Most of these terms apparently originate in the studios. Perhaps in 
sending out this questionnaire it might be well to ask the studios to add any new 
terms they have recently concocted. It might be well to split the terms into 
groups, each applying to its particular branch of the industry. This will reduce 
the number of terms for each branch to consider after receiving the questionnaire 
and therefore insure more prompt reply. Also, in addition to a dictionary ar- 
rangement of all the terms, it might be well to list the terms according to the 
various branches of the industry. 

MR. MORGAN: I suggest, in addition to sending the questionnaire to indivi- 
duals, that it be sent to the various department heads, such as the sound labora- 
tory and camera departments. If they could not find the time to do it themselves, 
they would probably delegate someone to prepare the answers. 

MR. ROBERTS: Has the Committee considered standards of tolerance for 
sound-track placement? 

DR. CARVER: Standards of tolerance for sound-track placement were in the 
old standards, and are in the proposed standards. The tolerance is 0.003 inch. 

MR. ROBERTS : I understand that the Academy has considered a 2-mil stand- 

DR. CARVER : The Academy is doing quite a lot of work on the problem, and we 
have appointed a sub-committee on the same subject. Mr. Davee is the Chair- 

MR. ROBERTS: I am quite sure that they are working toward ==2 mils, but I do 
not like to see that standard at present. The Academy proposes to give quite a 
bit of tolerance to projector scanning-beam width and to projector film-path 
misalignment. These two projector tolerances will be larger than the combined 
tolerance of negative and print. In a print it will be necessary to squeeze the 
combined misalignment of negative and print into =*= 0.002 inch. In the case of 
processes where dupes are made photographically, a recorder misalignment and 
three printing machine misalignments were all added together. These three can 
be kept to 0.003 but are extremely difficult to keep within ==0.002 inch. 

DR. CARVER: I think the answer to the tolerance question will have to come 
through a consideration of what the laboratories can maintain. With your new 
instrument they probably can do a lot better job in the future, and therefore the 
tolerance might be cut down. 


Summary. An indication is given of the quantity of illumination used on 
interior motion picture sets together with the technic of "key-lighting." Key-lighting 
levels in common use are stated and some of the variables in light balance are men- 

A photographic technic, previously described, 1 utilizing the photo- 
electric cell meter and the so-called system of "key-lighting" is 
rapidly gaining in popularity in the West Coast studios. 

The key-light or illumination falling upon the immediate fore- 
ground, particularly upon the faces of the dramatically important 
characters, is used as a middle-range basis upon which to balance 
the shadow and highlight areas of the set. This key-light is estab- 
lished visually by the cinematographer and checked with a photo- 
electric cell meter. 

Such a procedure, if followed correctly, produces a negative with 
the key-light exposure density midway between the underexposure 
and overexposure portions of the gamma curve, and therefore allows 
the greatest latitude for balance without blocked out shadows and 
burned up highlights. It also allows the laboratory to process the 
negative under time-and-temperature conditions with a reasonable 
assurance that the negative will print within a narrow range of printer 

In black-and-white photography no attempt is made to check the 
highlight and shadow areas with a meter. As the use of the meter 
becomes more common the shadow, key-light, and highlight areas 
may all be checked as a precautionary measure, but at present it is 
thought that the single measurement of key-light is sufficient to 
enable the cinematographer to establish the balance so there will be 
detail in both shadow and highlight areas if he wants it there. 

Certainly, on a new type of film, the H&D curve of the emulsion 
should be checked for latitude, and meter readings taken of shadow 
and highlight areas to determine whether or not the exposure lies 

* Presented at the 1939 Fall Meeting at New York, N. Y.; received October 6, 




within the straight-line portion of the gamma curve, in order to 
maintain density within the limits of correct exposure. 

The key-light levels on black-and-white sets, with the exception 
of extreme effects, range between 75 and 150 foot-candles, depending 
upon the lens aperture used, the individual technic of production, 
and the laboratory processing methods of the various studios. 

However, in analyzing data on set lighting, emphasis should be 
placed upon the fact that the key-light level is only an indication 

FIG. 1. Black-and-white: key-light level, 140 foot-candles. 
(Courtesy Paramount Pictures, Inc.) 

of the total illumination used. Measurement of the key-light is 
usually made with the meter held close beside the subject and the 
light-sensitive cell pointed at the key-light source and receiving the 
rays that would normally fall upon the face of the subject. Con- 
siderable judgment is exercised in the exact positioning of the light- 
sensitive cell. 

If there is little photographic contrast between the character's 
costume and the walls of the set it may be necessary to interpose 
backlight from the parallels of the set, between the character and 
the walls, of the order of from two to three times the intensity of 
the key-light level to create the illusion of depth. Shafts of light, 


such as sunlight and shadow effects, require several times the in- 
tensity of the key-light. Also, certain types of sets call for high 
levels of illumination while others are dark and dreary although the 
key-light may be the same. 

In Technicolor, or any color process, the addition of color precludes 
the practicability of direct quantitative comparison between the 
total illumination used on a color set and a black-and-white set. 
The key-light levels used with Technicolor vary from 150 to 400 foot- 

FIG. 2. Technicolor: key-light level, 300 foot-candles; left 
side highlight, 500 foot-candles; right side light, 250 foot -candles. 
(Courtesy Paramount Pictures, Inc.) 

candles, and sometimes higher, yet the total light on a Technicolor 
set with a key-light level of 400 foot-candles can not be compared 
directly with a black-and-white set with 150 foot-candles key-light 
level because it may not be necessary in the case of color to separate 
the characters from the background with a wall of intense light. 

Also, in black-and-white photography, the exact shade of gray in 
which a costume is rendered is of relatively small importance, whereas 
in color pictures the total effect is enhanced by the faithfulness of 
reproduction of the original costume. 

As an example, a deep green velvet dress with many folds may 


photograph dark gray in black-and-white photography and the 
shadow areas may be totally dark, but the effect may be acceptable. 
In color, however, it is necessary to illuminate the shadows so the 
final print will reveal a green dress with green, rather than black, folds. 
In dealing with black and white, where the absence of light means 
black, it is often possible to ignore deep shadow areas without at- 
tracting the attention of the casual observer. But in color, where the 
absence of light also means black, it is essential that the shadow areas 

FIG. 3. Showing position of photoelectric cell meter in ob- 
taining key-light; Cinematographer Victor Milner. (Courtesy 
Paramount Pictures, Inc.) 

be illuminated with sufficient intensity to give at least a suggestion of 
their true color. 

C. W. HANDLEY, Chairman 




1 CLARK, D. B. : "Methods of Using and Coordinating Photoelectric Exposure- 
Meters at The 20th Century-Fox Studio," /. Soc. Mot. Pict. Eng., XXXIII (Aug., 
1939), p. 185. 





Summary. Working toward closer coordination between the creative and 
technical groups in the making of motion pictures the author points out the objective 
and factors essential to good motion picture entertainment. Subsequently to the 
analysis of the general problem, examples are given illustrating the types of problems 
where technical knowledge can assist in obtaining a better dramatic result. 

The information presented in this paper has been taken from a 
paper which the writer prepared at the request of the Paramount ad- 
ministration for presentation to the producers, directors, and writers 
of our company. The thought is that too large a gap exists between 
the creative and technical groups, and that this and other discussions 
of the overlapping problems between our creative and technical 
groups will make for better motion picture entertainment. 

The producers, directors, and writers were asked to consider the 
problem from the standpoint of the technician and in this case I am 
going to ask you to consider the problem from the standpoint of a 
producer or studio executive. Your job is to create entertainment. 
Let's stop and analyze the problem. Motion pictures are a means of 
expression. As defined in the Britannica it is a means of ' 'transmitting 
emotional stimuli and experience by re-creating events." It is the 
conveyance of an illusion to the audience. 

For the most part we are letting our audience "listen in" on the 
intimate lifelike happenings of our characters. But, at times we want 
our audience to feel that they are a part of the dramatic story which 
is being told, as for example, the earthquake in the picture San 
Francisco. Good motion picture entertainment takes the attention of 
the audience from their thought of every-day activity, and even away 
from a realization that they are in a theater. Then after so captivat- 
ing the interest of the audience we carry them around from place to 

* Presented at the 1939 Fall Meeting at New York, N. Y.; received October 
12, 1939. 

** Paramount Studios, Hollywood, Calif. 



place, play on their emotions with music, give them realism with 
effects, and swing them into a climaxing end title. 

Most of our effort in the past has been toward technical improve- 
ments many of which have heightened the entertainment value of our 
pictures. But we are approaching a point where the technical status 
of our work is ahead of our dramatic usage of technical knowledge and 

In sound, just as in pictures, we have never reached that success of 
complete illusion, and, we ask ourselves, Why? Naturally, we realize 
that our means of expression is at the moment limited by a flat screen 
for the picture and a point source back of the screen for the origin of 
sound. In attempting to use this limited means of expression, we in 
the motion picture industry have worked out a technic which motion 
picture audiences have learned to accept. But, is this the best set-up 
and are these the best technics? And the answer is only if it gives 
the best illusion and the best entertainment. 

My objective is not to revolutionize this business, but to point out 
more definitely our objective, the objective of the producers and the 
studio executives and the types of problems where technical knowl- 
edge can assist in obtaining a better dramatic result. We will start 
by considering certain of the problems for which we have no direct 
and complete answer. 

Much of the illusion of intimacy is lost in our dramatic and love 
scenes. Pictorially we bring our characters closer to the audience by 
using closer shots, but from the standpoint of sound we find ourselves 
compromising. Increased levels play against the illusion of intimacy. 
And, if we reduce the volume appreciably, the sound will feel artificial 
against the large picture. In shooting the scenes our microphone is 
placed close for an intimate pick-up, we use "effort equalization" 1 
or the equivalent, but the audience distance from the screen plus the 
theater reverberation prevent accomplishment of the desired effect. 
What we want is the effect of the actor speaking low but close to the 
listener the effect obtainable when you are seated close to your radio. 

I will avoid a discussion on theater acoustics and speaker equip- 
ment, but some day someone will supply the answer; it may be as 
simple as a new speaker system directional at all frequencies and 
directed at the audience, thus minimizing the distant effect caused by 
reverberation. Meanwhile, producers, directors, and writers are 
limited in the scope of story construction by our inability to remove 
such obstacles and our means of cooperation (as technicians) should 

100 L. L. RYDER [J. s. M. P. E. 

be by recognizing these problems and finding ways of meeting them. 

Shouts and high-volume commands frequently lose their force and 
become thin and undramatic as reproduced in the theater. This may 
be partly caused by the dialog bass suppression used in the recording 
channels. (The bass suppression is usually set for normal dialog, 
which is incorrect for a shout.) But we are safe in saying that most of 
this trouble is caused by overload either in the recording system or in 
the theater reproduction. 

We encounter still another problem when we consider dynamic 
expression. Our technic evolved from practice indicates a necessity of 
playing normal dialog at an increased level to the listener as compared 
to the sound level at the point of pick-up. The volume levels at 
which sounds become objectionable to the listener have not been 
changed, so by this practice we have reduced our permissable range 
of expression as compared to normal life. The result is less effective 
entertainment and the problem will probably not be answered until 
steps are taken to reduce our normal dialog playing levels in the 

We should ask also what is the level at which sound becomes 
objectionable to the listener? To this I can positively say that the 
limit of desirable volume in most theaters is dependent upon the 
power-carrying capacity of the reproducing equipment. Overload 
changes sounds of rounded true tonality into a conglomeration of 
harsh, piercing, and unpleasant sounds. Our organization has made 
many comparisons and our executives have observed the difference 
in audience reactions and acceptance of spectacle sequences. In 
theaters with ample power the effects seem real and dynamic, while 
in houses equipped with insufficient power the effect of the sequence 
is lost and the sound seems objectionably high even though the 
acoustical energy has been reduced by the over load. 

Volume range is also limited in the downward direction by noise, 
but to this end we are making progress as reported in Dr. J. G. 
Frayne's paper "Report of Progress on Adaptation of Fine-Grain 
Film to Variable-Density Sound Technic" 2 and Dr. C. R. Daily's 
paper "Improvement in Sound and Picture Release Through the Use 
of Fine-Grain Film." 2 In this regard there are some observations 
worthy of note. At the preview of Geronimo (the first picture 
"movietoned" and shown to an audience on fine-grain film) the 
audience did not analyze, and probably most of them did not notice 
the reduction in film background noise, but their reactions, especially 


the complete quiet of the audience during dramatic sequences, was 
evidence enough for the studio executives. The writer and all those 
present were convinced of the value of reducing noise, especially 
noise from the screen. 

It is important to note that sound has progressed to the point 
where improvements are no longer evaluated by direct audience 
criticism or comment but none the less surely by the audience reaction. 

Music is a most important factor in screen expression but few 
persons realize the complexity of technics in music handling in con- 
junction with picture shooting. In a recent study we found seventy- 
eight different combinations for handling music involving pre-scor- 
ing, separate vocal and orchestra recording, direct recording, cueing, 
playing back, and post recording. In fact, at times so much thought 
is given to the mechanics of tying musical recording into the picture 
shooting that the musical objective is forgotten. Music is used for 
creating an emotion within the audience, as in an under-score ; giving 
a complete concert rendition as the Damrosch number in Star Maker, 
establishing tempo in a dance sequence, punctuating action, or 
"plugging" a number with the hope of making a hit tune. 

There are many problems and shortcomings in meeting these 
demands. For instance, in a tap-dance routine, the music is usually 
recorded first, a "pre-score," so as to establish a tempo guide which 
can be played back or reproduced on the production stage and used 
as a guide for shooting all the scenes and angles of the sequences in 
their various camera angles. If the "playback" is reproduced low so 
as to obtain a good recording of the taps, our dancer has trouble 
keeping in step with the music or in any case his action tends to drop 
with the music level and we do not obtain the desired punch for the 
scene. If, on the other hand, we "play back" at a higher level we 
either override the taps with music or in any case pick up too much 
of the "playback" music for subsequent handling in dubbing. Fre-. 
quently we "post-score" the taps to pictures and guide track subse- 
quent to editing of the picture and the result is an out-of-sync, none- 
too-realistic sequence. The same general problem is encountered in 
much of our vocal work and I know you have all seen and heard good 
scenes and good music made unconvincing because of poor syn- 
chronization. A solution might be a directional speaker and a direc- 
tional microphone arranged so as not to work into one another or 
possibly a phasing device to cancel the picked-up noise of the music 

102 L. L. RYDER 

Still striving toward our objective of better entertainment we can 
find more problems affecting our re-recording or dubbing, our proc- 
essing and our release, but these are all problems based on our present 

Sound has in a way built a fence around itself. Our past limitations 
are being used by the entertainment creators as a guide of future 
possibilities. They, the creators, have no way of expanding the scope 
of picture entertainment unless our vision reaches out, obtains the 
tool, and presents them to the creators for incorporation in a picture. 
We should study all forms of entertainment and borrow, if you 
please, any technic or equipment set-up which will aid in picture 

We are not without recent developments of this type. The Bell 
Laboratories have presented the vocoder and Gilbert Wright of 
Hollywood has demonstrated a new re-modulating device. 

Showmanship is a peculiar business and our audiences are fre- 
quently hard to entertain, but there is one thing certain and that is 
that audiences enjoy something new, something different. They 
always react to things that they have never seen or heard before. 


1 MORGAN, K. F., AND LOYE, D. P.: "Sound Picture Recording and Repro- 
ducing Characteristics," /. Soc. Mot. Pict. Eng., XXXIII (July, 1939), p. 107. 
1 Pp. 3 and 12, this issue of the JOURNAL. 


During the Conventions of the Society, symposiums on new motion picture ap- 
paratus are held, in which various manufacturers of equipment describe and demon- 
strate their new products and developments. Some of this equipment is described 
in the following pages; the remainder will be published in subsequent issues of the 


Present-day sound recording for motion pictures generally takes place under 
one of the following conditions: 

(1) Original recording on a studio stage. 

(2) Original recording on location. 
(5) Re-recording. 

(4) Background projection. 

Up to the present time no one motor system has been designed which will 
operate satisfactorily under all these circumstances. The synchronous motor 
system is convenient on studio stages but requires an accurately regulated a-c 
supply which is often not available on location. The d-c interlock system pro- 
vides adequate portability for location work but requires the use of portable stor- 
age-batteries which are undesirable on the stage. Neither of these two systems 
provides interlock from start and, therefore, can not be used for re-recording or 
background projection for which an a-c interlock or Selsyn system must be used. 

The new multiduty motor system not only provides a single system for all these 
purposes but also better operation in its various applications. One of the first 
features to appeal to the studio sound and camera departments is that the same 
motor can be used on the camera whether shooting at the studio or on location. 
The same is true of portable recorders which may be used either in the studio or 
on location without change of motors or operating technic. At the same time the 
new system provides more power for camera motors without increase in size, more 
accurate interlock, and a considerable number of accessory features which add 
materially to the convenience and reliability of operation. 

Multiduty Motors. The basis of the new system is the development of a new 
type of motor which is capable of operating satisfactorily on either a-c or d-c 
power supply. In Fig. 1 it will be noted that the multiduty motors are basically 

* Presented at the 1939 Fall Meeting at New York, N. Y.; received October 
12, 1939. 

** Electrical Research Products, Inc., Hollywood, Calif. 




d-c motors with the commutators tapped for 3-phase interlock as used in previous 
d-c interlock systems. D-c power, in this case from a common 96-volt battery 
or generator, supplies field and armature current for each motor. When power is 
applied to the individual motors they each start as d-c shunt-wound machines. 
As soon as d-c is applied to the armature, a voltage appears at the slip-rings 
which becomes an alternating three-phase potential as the motor rotates and 
provides the necessary interlock power. A relatively large amount of short- 
circuited damper copper is distributed through the d-c field structure in such a 
form as to provide a squirrel-cage winding. 

When these motors operate in a d-c interlock system a difficulty occurring in 
some previous systems, due to short-circuiting through the interlock windings 
when the motors are out of phase, has been overcome by including ballast resistors 
in series with the interlock connections. These resistors have relatively low 
resistance when cold but increase to roughly 12 times this resistance when called 
on to carry short-circuit current. In operation the motors, when in phase with 
each other, interchange relatively little current, and the resistors, being cold, are 


FIG. 1. Multiduty motor system. 

of low impedance. Under a short-circuit, however, this resistance builds up with 
sufficient rapidity to protect the motors against destructive current interchange. 

Another desirable feature of the ballast-lamp resistors is that they provide a 
pilot indication of interlock condition, showing an increasing glow as the motors 
tend to pull apart, which becomes a bright flickering when the motors actually 
pull out of phase. This permits the operator to adjust the normal speed of each 
motor quickly and with sufficient accuracy to bring an out-of-line motor into step 
at the beginning of a take, while under previous methods of operation it was neces- 
sary to stop the take and recheck motor speeds independently. 

The motors when operating as d-c units are essentially d-c motors, delivering 
mechanical power to the shaft, and inverted converters delivering any necessary 
power to the interlock circuit. When operating from a 96-volt d-c power supply 
the camera motor produces approximately 300 mechanical watts and operates at 
about 50 per cent efficiency. 

Fig. 2 shows another totally different method of operation which requires no 
change in the motors. In this mode three-phase alternating-current is supplied 
to the rotor through the slip-rings which were used for interlock when operating 
on d-c. No ballast-resistors are necessary in the three-phase connections be- 
tween rotors since the d-c supply to each motor has been opened. The winding 

Jan., 1940] 



which has been functioning as a d-c armature now becomes, in effect, a three- 
phase delta-connected winding, and when power is supplied to this winding, it 
reacts with the damper copper distributed through the pole-pieces and provides 
induction motor torque sufficient to start the motor and bring it close to syn- 
chronous speed in the same manner as that employed with the induction-syn- 
chronous motor. The motor will pull into synchronism in much the same way 
because of the salient poles which are in the stator in this case instead of in the 
rotor as in the usual synchronous motor. 

The amount of copper which it is possible to dispose throughout the stator 
while still maintaining adequate space for the d-c field winding is not enough to 
provide sufficient power for a camera motor without increasing the size beyond 
desirable bounds. However, it has been found that by proper poling of the d-c 
field windings they can be made to assist the induced poles of the squirrel-cage or 
damper winding if the field winding is self-excited from its own commutator. 
This assistance can be made to provide somewhat more than twice the total 



FIG. 2. Method of operation requiring no change in the motors. 

power available from the squirrel-cage. Thus in the camera motors of the multi- 
duty system, the squirrel-cage copper alone will synchronize approximately 85 
watts of mechanical power with a normal applied voltage. The addition of the 
self-excited field windings increases this maximum pull-in to over 200 watts, 
which is somewhat greater than the power supplied by any camera motors now in 
the field except those designed for Technicolor cameras. An added advantage of 
the self -excitation is the fact that the power-factor of the primary or three-phase 
winding may be brought up to any desired value and this, in turn, means low 
copper losses and relatively high efficiency. 

When operating from 65-volt, three-phase, 60-cycle supply these motors have 
an output of 200 watts with an efficiency of approximately 50 per cent. The 
a-c supply is usually obtained from a three-phase autotransformer associated 
with the central control unit. 

A third mode of operation is shown in Fig. 3, which differs from the synchronous 
mode in two respects. The windings are not self-excited but are supplied from 
an external d-c source which permits them to be fully excited at standstill. The 
three-phase supply is from an inverted converter driven from 110 volts d-c in- 
stead of a supply-line as for synchronous operation. While running, the motors 



are essentially synchronous motors as in Fig. 2, except that the speed may be 
varied by varying the speed of the inverter which supplies them. Thus the 
motors will run at any desired speed, depending on the speed of the control motor, 
from zero to considerably above normal. At zero speed the inverter is, of neces- 
sity, at a standstill and of course if full voltage is still applied to the d-c input to 
the inverter, the current and heating will become destructive. However, if the 
d-c voltage is reduced, either by tapping down on a supply -battery or introducing 
series resistance, this heating may be held within acceptable limits for short periods. 
At the same time sufficient voltage is provided on the three-phase leads to the 
driven motors to maintain adequate current in the rotor windings so that the 
motors still remain aligned with their externally excited fields. The required 
voltage is very much less at standstill because of the fact that the impedance of 
the motor windings at standstill is reduced to resistance only. Since the distribu- 
tion of voltage applied to the three-phase windings is determined by the position 


FIG. 3. A third mode of operation. 

of the inverter rotor under its own brushes, the motors are effectively aligned with 
it and with each other at standstill and will come up to speed in the same align- 
ment. A system utilizing this feature was used for some time at Universal 
Studio some years ago and proved the basic theory of operation to be sound. 

This system has the disadvantage of requiring the inverter distributor which 
would seldom operate at as much as 50 per cent efficiency and which constitutes 
considerable weight in itself. With the increased batteries this system will weigh 
about four times as much as would a similar d-c interlock system without inter- 
lock from start. However, the new system is considerably lighter and less bulky 
than a Selsyn system. 

The system outlined above is capable of operating in all three of the desired 
modes : d-c interlock for location work in which a high order of portability is de- 
sirable, synchronous for stage use, and interlock from start for special duties which 
may be required. In addition to the ability to operate in all the desired modes, 
these motors have the advantage of providing a coupling with each other, or with 
a supply-line, which is much tighter than any other system. This is due to the 

Jan., 1940] 



fact that the motors have salient poles which are also excited by a d-c winding. 
The magnetic circuits are thus much more sharply defined than is possible with 
other types of synchronous or interlock motors. This rigidity of coupling can be 
measured by observing the position of a stroboscopic image while a motor is run- 
ning "no load" and then applying some known percentage of the full load and 
observing the displacement of the image. Motors of various types can thus be 
compared provided shaft speeds are the same. The multiduty motors show 
about one-half the displacement of variable-reluctance synchronous motors and 
about one-fourth the displacement of the a-c interlock system. This factor is of 
chief importance in background projection where the phase relation of camera 
shutters to the projector shutter determines the exposure of the picture negative, 
and the new system should provide a considerable improvement for this duty. 

FIG. 4. Portable control cabinet. 

Other System Features. With a type of motor available which is capable of all 
these modes of operation it was felt desirable to put them into operation in a 
manner which would serve the needs of the greatest number of producers. Con- 
sequently, with this in mind, a survey of the studios was conducted to determine 
the best method of adapting this system to their varied uses. As many as possible 
of the features desired by the studios have been incorporated in the auxiliary 
equipment described below. 

A-c-D-c Control Cabinets. Most of the studios desired some form of centralized 
control of all motors, whether they were operating on d-c or on a-c. This has been 
provided in the form of a portable control cabinet (Fig. 4) which serves for start- 
ing the motors and controlling their speed when operating on d-c, indicating the 
operation of the system, and other functions which will be described below. 


Provision is made for connecting the various motors to the control circuits by 
means of standard six-conductor plugs, jacks, and cables. In order to provide 
the desired auxiliary features, it was found necessary to increase the number of 
conductors above the six required for actual motor operation. These auxiliary 
functions were combined in a second cable which was a standard six-conductor 
No. 18 speech cable, making it possible to utilize surplus microphone cables for 
this purpose. Provision is made for connections to three cameras and one film 
recorder. If this should prove insufficient for some special purpose, a second con- 
trol cabinet may be connected in multiple with the first cabinet, making provision 
for four more motors. 

Various desirable features, such as ability to slate independently or operate 
from a central source, and the assurance that slating will not have left motors off 
the line, are provided for by the use of relays for controlling motor circuits. 
Thus for each motor there is an a-c disconnect relay, an a-c "synch" light relay, 
and an a-c slow-start relay; for d-c operation, a d-c disconnect relay and a d-c 

FIG. 5. Switching and junction box. 

slow-start relay, thus providing a total of five relays per motor. Since approxi- 
mately one-half of these relays are idle on any given mode of operation, the com- 
plication involved is more apparent than real. From a maintenance standpoint, 
relays of the type used are to be preferred to switches since they lend themselves 
to ready inspection and are purposely mounted with the contacts immediately 
accessible for any necessary cleaning operation. Any burning or blackening of 
the contacts can be observed and corrected, whereas with switches such a burning 
would only make itself known by switch failure. 

Independent "Slating" for Cameras. A small switching and junction box is 
provided for each camera motor (Fig. 5). This is attached to the camera and 
provides a termination for the two cables from the central control box. It is so 
arranged that the cameraman may run the camera independently of the rest of 
the system by simply throwing a switch. This is of particular value for "slating" 
and "wild shots." As noted below, a pilot-light indicates to the recording opera- 
tor when the camera is being run, allowing him to check, and, if necessary, correct 
the speed at that time. 


Automatic Slow-Speed Start. Since the motors are designed to supply sufficient 
power for the heaviest loads, it was found desirable to incorporate a starting con- 
trol so that light loads would not be started too abruptly. This was done by 
including in both a-c and d-c supply-circuits adjustable resistances which are 
shorted by relays whose coils are excited from the armature circuit. In this way 
the individual circuits may be adjusted so that all motors come up to speed with- 
out jerking and in approximately the same time. 

A-c-D-c Running Pilot-Lights. A pilot-light is provided for each motor circuit 
to indicate when the motor is running. When running on d-c, this permits the 
operator to adjust the motor speed without running the motor especially for this 
purpose. It is useful in a-c operation to show which motor is running but not 

Synchronizing Pilot-Light. A red "synch" light is used for a-c operation and 
acts as a pilot-light to indicate whether the three-phase supply is connected to 
the control cabinet. The circuit is arranged so that this light goes out when the 
last machine is synchronized with the line. This also serves to indicate if a motor 
drops out of synchronism for any reason such as poor contacts in motor cables. 
On d-c operation the ballast resistors in the interlock circuits serve a similar pur- 
pose. These flash for "out-of-synch" conditions, but are dark when the motors 
are properly interlocked. These indications of synchronism are useful to the 
operator in order that he may make the synchronizing mark or "bloop" with as 
little delay as possible and yet be sure that all motors are actually in synchronism 
when the mark is applied. 

Synchronizing or "Bloop" Marks. The synchronizing circuit is manually oper- 
ated by the operator from the central control cabinet. The "bloop" switch is a 
momentary contact, push-button type, and marks the picture film by applying 
an alternating voltage to an argon lamp in the camera which is in series with a 
buzzer at each camera. This operation also marks the sound-track by applying 
the same signal to the light-valve. A pilot-light in the cabinet is in series with the 
buzzer at each camera and the marking light inside the camera. Thus the pilot- 
light glows dimly when the circuit is normal, does not light at all when the circuit 
is open and glows brilliantly if the light in the camera is shorted. The buzzer at 
the camera furnishes a signal to the set that the motor has reached normal speed 
and is synchronized. This permits the action to start more quickly than when it 
is necessary to signal back and forth between stage and recorder position. This 
circuit has also the advantage that it is a completely separate circuit and is less 
likely to interfere with other circuits or to be interfered with than where it is 
"phantomed" on circuits designed for some other purpose. 

Speed Indication and Adjustment. Since the d-c system is usually used without 
a master speed-control in order to obtain maximum portability, it is necessary to 
have an accurate speed indicator to show when the proper speed is reached. This 
is provided by means of a dual-range tachometer with one scale covering the range 
between 82 and 93 feet per minute. It is also desirable to be able to operate the 
system at less than normal speed for special effects. The lower speed range is 
obtained with the d-c system by tapping the supply-battery down to one-half 
normal voltage in 12-volt steps. This provides stability at lower speeds compar- 
able to that obtained at normal speed. The speed in this range is read in frames 
per second on a second scale of the tachometer. A voltmeter is also provided for 


reading battery voltage as supplied through the tap switch and therefore reduces 
the likelihood of the tap being left on a subnormal position. 

Film-Buckle Release. The film-buckle release switch at each camera carries 
only relay-coil current and thus is protected against excessive burning and arcing 
common to most other systems. This switch may be used as a disconnect switch 
by cameramen without the previous penalty of high maintenance. 

Other Control Cabinets. While it was originally intended that the a-c-d-c 
cabinet would serve for either stage or location duty, it has become apparent that 
some studios have such a high percentage of stage operation that the d-c equip- 
ment in most a-c-d-c boxes would have very little use. To meet this condition a 
cabinet has been designed for non-portable stage use only, which contains only 
those elements necessary to a-c operation. While a box for use only on d-c has 
been schematically designed, it has been generally contended that for portable 
use the a-c-d-c box is desirable because of the fact that many locations would 
permit operation on a-c even though they may be many miles from home. A 
control-box has been designed to operate any type of standard synchronous motor 
which might be used on a recorder in conjunction with the new system. This unit 
contains the necessary units to start and "bloop" the whole system when it is 
desired that the recorder man assume this function and will operate either in con- 
junction with the a-c only or the a-c-d-c box. 

Background Projection. For background projection a cabinet is provided 
which will work in conjunction with either the a-c-d-c cabinet or the a-c only 
unit on either a-c or d-c supply. The auxiliary cables from each motor go first 
to this background projection cabinet and then to the main control cabinet via 
short jumpers. A contact at each motor which is mechanically coupled to the 
shutter effectively indicates to the central circuit when the shutters of each ma- 
chine are open. The control circuit then compares the shutter position of each 
camera with that of the projector and automatically adjusts the shutter positions 
so that all are open at the same time by momentarily disconnecting any whose 
position is not correct and dropping it back until such time as the shutter does 
align. A visual indication of the phase relations of the shutters is provided in 
this cabinet so that the operator may observe the actual functioning of the auto- 
matic circuit and, if necessary, phase in the various units manually by operating 
the associated disconnect switches. This is intended to provide against possible 
failure at an inopportune time of the automatic facilities and it also provides a 
means which is not available in any present system of determining that the 
shutters are in proper alignment throughout the take. This method of shutter 
phasing has the further advantage that it will work equally well on either a-c or 
d-c and consumes practically no additional power. It seems possible that due to 
the reduction in power consumption and its associated reduction in weight and 
bulk, this feature may make feasible background projection on distant locations 
which have not been considered practicable before. 

It has been the aim throughout the development of this system to include only 
features that a majority of the studios were agreed upon as desirable, but at the 
same time provision has been made in many cases to provide facilities that can be 
added to meet special needs. For instance, automatic speed control does not seem 
to be justified for location work and the multiduty system obtains speed control 
from the supply-line for stage use, but if sufficient demand for this feature arises, 


it can be added without altering existing equipment. The motor speeds have 
been based on the prevalence of a 60-cycle supply frequency but the motors can 
readily be wound to operate on 48 cycles where a 1440-rpm speed is essential for 
direct shutter-shaft drive. Possibly some of the many facilities and indicators 
that have been included will prove non-essential in the field and it is possible that 
some essential items have been omitted. However, this is the first time that a 
motor-system has been made available in which the auxiliary features necessary 
to smooth, efficient action on the set have been incorporated as an inherent part 
of the system. Considerable direct saving in both time and wasted film should re- 
sult from the use of these facilities and a further indirect saving should be realized 
from the fact that a wide range of operating conditions can be met without a 
material change of either equipment or operating technic. 

It is desired to thank those in the industry who have helped define these re- 
quirements, and particularly Paramount for the field tests and the suggestions 
contributed by the sound department staff. 


MR. READ: You stated in the paper that the stiffness of coupling or deflection 
angle for a given load was half for the motor with a d-c field over that with the 
induced field, or variable-reluctance motor. With the motor with the d-c field 
you would expect to reach maximum torque at 90 electrical degrees deflection. 
With the other type of motor, you would have to obtain maximum torque at a 
smaller value than that, because at 90 degrees it would have passed the point of 
stability and be slipping a pole. 

MR. KEITH: I believe the first part of your question referred to the difference 
in the coupling between the d-c and the a-c motors. 

MR. READ : I referred to your using both motors as a-c motors in one case 
using the conventional squirrel-cage motor, which has part of the rotor cut 
away, to make it pull into synchronism; and in the other case a motor with a 
d-c field. I understood that the one with the d-c field had twice the stiffness 
of coupling. 

MR. KEITH: The one with the additional self -excited d-c field has greater 
coupling than the one with a simple synchronous motor. 

MR. READ: Assuming that each type of motor is designed for the same 
maximum pull-out torque, I would appreciate an explanation of why the self- 
excited d-c field has greater stiffness than the simple synchronous type. 

MR. KEITH: I believe the difference is the distribution of flux about the poles 
rather than in the difference in form of operation. 

MR. HOLCOMB:* Mr. Keith is correct in that the sharper coupling is due to a 
difference in flux pattern. In practice the variable-reluctance synchronous camera 
motor operates in synchronism essentially as if the rotor were a bar of iron which 
follows the rotating flux of the stator and seeks to occupy the position of greatest 
flux density at all times. The angle of lag introduced by any given load on the 
rotor will depend on the shape and coverage of the rotor poles. Thus one such 
rotor might have the area between poles relieved so that the air-gap increased 

* Communicated. 


uniformly from the center of the poles; another rotor might be slotted abruptly 
and deeply between poles. Both rotors might have the same maximum torque 
or pull-out, but a very different response in angular position to changing loads at 
less than pull-out torques. The power of any type of synchronous motor that 
must start under full-load conditions is determined by the load which it can pull 
into synchronism, and this value is usually much less than the motor will maintain 
when once in step. This fact, added to the requirement that camera motors 
operate with a considerable margin of safety, with respect to power, means that 
the average motor works at less than 20 degrees displacement, and thus never 
approaches the theoretical 90 degrees mentioned. The "pull-in" and mechanical 
considerations dictate to a large extent the slotting of most variable-reluctance 
synchronous motors rather than sharpness of coupling, and while this type of 
motor could be designed for a coupling characteristic equal to that of the multiduty 
type it probably would require a larger and heavier frame. The d-c excitation 
on the fields of the multiduty motors increases the flux density without ma- 
terially increasing the leakage, and thus improves, the definition of the flux pat- 

MR. KELLOGG: The point was made several times in the paper that the 
efficiency is very high. Are there any particular design features that would make 
it any more efficient, for example, then a d-c motor of about the same size? 

MR. KEITH: I do not believe it was meant that the motors, as constructed, 
are more efficient than a simple d-c motor without interlocking windings, but 
more efficient than other systems of d-c interlocked motors. 

MR. KELLOGG: According to the diagram, there was no extra winding. I 
have been under the impression that some motors now in use had a low-voltage 
d-c winding supplemented by a high-voltage a-c winding, making the machine 
like an a-c d-c dynamotor rather than a rotary converter. 

MR. KEITH: True; some systems have been used where the a-c interlock was 
obtained by a separate high-voltage winding where the a-c interlock circuit is 
separate from the d-c circuit. In these motors that is not done; the interlock 
circuit is taken directly from the commutator. 

MR. HOLCOMB:* The motor system to which Mr. Kellogg refers is the 12-volt 
d-c interlock system introduced by ERPI some years ago, which has a 12-volt 
d-c motor winding and separate 220-volt 3-phase interlock winding. In the 
multiduty motors the same rotor winding serves three purposes: (2) as a d-c 
motor winding, (2) as an interlock winding; (3) as a delta-connected 3-phase mo- 
tor winding. Since there is no "dead" copper in any mode of operation, the space- 
factor in the rotor is optimum. The stator, however, presents a difficult design 
problem due to the fact that both the d-c field winding and the short-circuited 
squirrel-cage winding must occupy essentially the same space or push the size of 
the motor beyond desirable limits. 

MR. KELLOGG: You refer to bars in the pole-pieces, so that you can start the 
machine as an induction motor? 

MR. KEITH: Yes. 

MR. READ: How does this machine differ in design from any staridard 3- 
phase rotary converter? Are there any features that would make it more efficient ? 
What is the size of the motor? 


MR. KEITH: I do not know the figures of the relative efficiency, compared with 
standard converter, but I see no reason why it would be any more efficient in any 
one feature. I think it is simply that they are more efficient than previously used 
motors, because of refinements in design and the number of small details. The 
camera motor is roughly 5 inches in diameter and 6 inches long, exclusive of the 
shaft extension. 

MR. KELLOGG: And that is good for 200 watts? 

MR. KEITH: 300 watts on d-c. 

MR. HOLCOMB:* The efficiency of the multiduty motors as stated by Mr. 
Keith is relative to existing systems intended for the same duty. The design 
problem is twofold: to produce a unit which delivers a high ratio of output to 
input, particularly for d-c operation work, where the size and weight of the power 
source must be kept to a minimum ; at the same time the size and weight of the 
camera motors must be held low. This is accomplished by careful design consider- 
ation of all losses, and a handmade type of construction which, while expensive, 
utilitizes all the available space in the motor to the best advantage. For operation 
on the stage as synchronous motors the efficiency is high because the power-factor 
is good, due to self -excitation, and the resultant low-current demand materially 
reduces the PR losses. 

MR. KELLOGG: What is the a-c voltage at which they operate? 

MR. KEITH: The supply is 65 volts a-c. 



The editors present for convenient reference a list of articles dealing with subjects 
cognate to motion picture engineering published in a number of selected journals. 
Photostatic copies may be obtained from the Library of Congress, Washington, D. C., 
or from the New York Public Library, New York, N. Y. Micro copies of articles 
in magazines that are available may be obtained from the Bibliofilm Service, Depart- 
ment of Agriculture, Washington, D. C. 

Acoustical Society of America, Journal 

11 (October, 1939), No. 2 
Normal Modes of Vibration in Room Acoustics: 

Experimental Investigations in Nonrectangular Enclo- 
sures (pp. 184-197) 
A Sound Source for Investigating Microphone Distortion 

(pp. 219-221) 

Microphone Efficiency : A Discussion and Proposed Defi- 
nition (pp. 222-224) 
The Degenerative Sound Analyzer (pp. 225-232) 

American Cinematographer 

20 (November, 1939), No. 11 

Fine-Grain Films Make Strong Advance (pp. 486-488) 
Sound Quality Improvements Obtained with Fine Grain 

Films (pp. 489-490) 

Making Modern Matte-Shots (pp. 493-495, 526), Pt. I 
Pacific Laboratories Announce Complete 16-mm. Service 

(pp. 497. 517) 
Studying Photoelectric Exposure Metering (pp. 499-500, 

524), Pt. I 

Eastman Issuing Two Classy Camera Models (p. 506) 
Here Are Tips on Editing and Splicing (pp. 508-509) 
Educating 300,000 with 16-Mm. Movies (pp. 510, 519) 
Densitometry and Its Application to Motion Picture Labo- 
ratory Practice (pp. 512, 513, 520), Pt. Ill 

British Journal of Photography 

86 (September 22, 1939), No. 4142 
Progress in Color (p. 587) 

86 (September 29, 1939), No. 4143 
Progress in Color (pp. 603-604) 

86 (October 6, 1939), No. 4144 
Progress in Color (pp. 611-612) 











86 (October 20, 1939), No. 4146 
Progress in Color (pp. 638-639) 

86 (October 27, 1939), No. 4147 
Progress in Color (pp. 647-648) 

British Kinematograph Society, Journal 
2 (October, 1939), No. 4 

Photographic Light Filters (pp. 215-222) G. J. CRAIG 

The Appraisement of Carbons for the Illumination of 
Kinema Screens (pp. 226-234) F. S. HAWKINS 

Educational Screen 

18 (October, 1939), No. 8 
Motion Pictures Not for Theaters (pp. 284-288) A. E. Knows 

International Photographer 

11 (October, 1939), No. 9 
Fine-Grain Release Prints (pp. 5-8) 
Mixing Weights and Measures (p. 10), Pt. 2 DON HOOPER 

International Projectionist 

14 (September, 1939), No. 8 

Ashcraft "Cyclex" Projection System (pp. 7-8, 10) J. J. FINN 

Flicker in Motion Pictures (pp. 18-20, 24-26) L. D. GRIGNON 

14 (October, 1939), No. 9 
Design and Operating Data on the "Cyclex" Projection 

system (pp. 7-14) D. S. ASHCRAFT 

Color Film Screen Values (pp. 16, 25-27) W. C. HARCUS 

Technicolor Adventures in Cinemaland (pp. 21-24) H. T. KALMUS 


21 (September, 1939), No. 9 

Beziehungen Zwischen Bild- und Tonsensitometrie (Rela- 
tion between Image and Sound Sensitometry) (pp 223- 
227) A. NARATH 

Die Entwicklung der deutschen Kinoprojektoren (The 
Development of German Motion Picture Projectors) 
(pp. 228-231) H. FICHTNER 

Die Kinotechnik in der Patentstatistik, 1938 (Motion 

Pictures hi Patent Statistics for 1938) (pp. 231-232) F. EARTH 

Photographische Industrie 

37 (September 20, 1939), No. 38 

Optische Kontrolle der Tonkopie (Optical of Sound 
Prints) (pp. 1039-1040) 

37 (September 27, 1939), No. 39 

Ein Vorschlag zum photographischen Aufzeichnen von 
Tonen aus dem Jahre 1880 (A Proposal for Photographic 
Sound Recording in the Year 1880) (p. 1055) 


Applied Acoustics; Harry F. Olson and Frank Massa; P. Blankiston's Son& Co. 
Inc., Philadelphia, Pa. (March, 1939). 

Although the subject of acoustics is very old, the early publications are devoted 
principally to scientific explanation of the phenomena observed. Early work on 
apparatus for production of sounds was limited almost entirely to the field of 
musical instruments. The invention of the telephone and later the vacuum-tube 
amplifier opened the field of modern acoustics. Means for recording, trans- 
mitting, and reproducing sound shifted the emphasis on acoustics from its aca- 
demic to its economic aspects. There followed a period of rapid development 
from which emerged industries of the first magnitude having their origin in applied 

This application of the science of acoustics is well treated in the book under re- 
view. The authors have had intimate contact with all the modern developments 
and have made many original contributions thereto. The early chapters give 
sufficient discussion of fundamentals to serve as a guide to experimental work and 
a check on results. The greater portion of the book, however, is devoted to a de- 
scription of apparatus and technic which have been developed during the past two 
decades for use in the fields of radio broadcasting and sound pictures. The de- 
scriptive matter is written so as to be understandable without thorough mastery 
of the theoretical work. The apparatus is discussed both from the standpoint of 
the designer and the user. 

The second edition is justified in view of the rapid strides which have been made 
recently in the development of microphones and loud speakers. While the prin- 
cipal changes from the earlier edition are in the discussions of these elements, 
changes in matter and presentation are made throughout the book where necessary 
to bring it up to date. It should be helpful to anyone desiring to become familiar 
with the latest developments in the field of applied acoustics. 






Officers and Committees in Charge 

E. A. WILLIFORD, President 
S. K. WOLF, Past-President 
W. C. KUNZMANN, Convention Vice-President 
J. I. CRABTREE, Editorial Vice-President 
S. HARRIS, Chairman, Papers Committee 
J. HABER, Chairman, Publicity Committee 
H. GRIFFIN, Chairman, Convention Projection 
E. R. GEIB, Chairman, Membership Committee 
H. BLUMBERG, Chairman, Local Arrangements 

Reception and Local Arrangements 






Registration and Information 

W. C. KUNZMANN, Chairman 


Hotel and Transportation 

E. O. WILSCHKE, Chairman 





J. HABER, Chairman 



Convention Projection 

H. GRIFFIN, Chairman 




Officers and Members Projectionists Local 310, IATSE 


118 1940 SPRING CONVENTION [j. s. M. p. E. 

Banquet and Dance 

M. C. BATSEL, Chairman 




Ladies Reception Committee 

MRS. O. F. NEU, Hostess 

assisted by 




Miss L. A. MOVER, Hostess, Chalfonte-Haddon Hall 


Headquarters. The headquarters of the Convention will be the Chalfonte- 
Haddon Hall, where excellent accommodations have been assured, and a recep- 
tion suite will be provided for the Ladies' Committee. 

Reservations. Early in March room reservation cards will be mailed to mem- 
bers of the Society. These cards should be returned as promptly as possible 
in order to be assured of satisfactory accommodations. 

Hotel rates. Special rates have been guaranteed by the Chalfonte-Haddon 
Hall to SMPE delegates and their guests. These rates, European plan, will be 
as follows: 

Four Lower 

Floors Ocean View Ocean Front 

Room for one person $3.50 $4.00 $5.00 

Room for two persons 6.00 7.00 8.00 

Parlor Suite, for one 10.00 12.00 14.00 

Parlor Suite, for two 14.00 16.00 18.00 

(All bathrooms at Haddon Hall have hot and cold running fresh and salt 


If American plan rates are desired the hotel room clerk should be advised 
accordingly when registering. An additional charge of $3 per day per person 
will be added to the above-listed European rates for three daily meals, American 
plan. Members and guests registering at the hotel on the American plan will 
pay only $3 for the SMPE banquet scheduled at Haddon Hall on Wednesday 
evening, April 24th. If registered on the American plan, the clerk at registration 
headquarters should be advised accordingly when procuring your banquet 

Parking. Parking accommodations will be available to those who motor to 
the Convention at the Chalfonte-Haddon Hall garage, at the rate of 50(f for day 
parking or $1.25 for twenty-four hours. These rates include pick-up and delivery 
of car. 

Registration. The registration and information headquarters will be located 
at the entrance of the Viking Room on the ballroom floor where the technical and 
business sessions will be held. All members and guests attending the Convention 

Jan., 1940] 1940 SPRING CONVENTION 119 

are expected to register and receive their badges and identification cards required 
for admission to all the sessions of the Convention, as well as to several motion 
picture theaters in the vicinity of the Hotel. 

Technical Sessions 

The technical sessions of the Convention will be held in the Viking Room of 
the Hotel. The Papers Committee plans to have a very attractive program on 
papers and presentations, the details of which will be published in a later issue 
of the JOURNAL. 

Luncheon and Banquet 

The usual informal get-together luncheon will be held in the Benjamin West 
Room of Haddon Hall on Monday, April 22nd, at 12:30 p.m. The forty-sixth 
Semi-Annual Banquet and Dance of the Society will occur on the evening of 
Wednesday, April 24th, in the Rutland Room of Haddon Hall an evening of 
dancing and entertainment for members and guests. 

Ladies' Program 

A specially attractive program for the ladies attending the Convention is being 
arranged by Mrs. O. F. Neu, Hostess, and the Ladies' Committee. A suite will 
be provided in the Hotel where the ladies will register and meet for the various 
events upon their program. Further details will be published in a succeeding 
issue of the JOURNAL. 


At the time of registering, passes will be issued to the delegates of the Con- 
vention admitting them to several motion picture theaters in the vicinity of the 
Hotel. The names of the theaters will be announced later. 

Atlantic City's boardwalk along the beach offers a great variety of interests, 
including many attractive shops and places of entertainment. 

Convention Vice-President 



As the result of the recent balloting for officers and managers of the Atlantic 
Coast Section, the successful candidates for 1940 are as follows: 

P. J. LARSEN, Chairman 
J. A. MAURER, Sec.-Treas. 
R. O. STROCK, Manager 

D. E. Hyndman continues on the Board of Managers as Past-Chairman, and 
the term of H. Griffin, Manager, has another year to run. 

Tentative arrangements for the next four meetings of the Section have been 
made as follows: 

January 10, 1940. Further discussion of high-quality 16-mm recording, and 
16-mm negative-positive and duplicate negative technic, by J. A. Maurer, of 
the Berndt-Maurer Corp., New York, N. Y. There is also the possibility of 
a paper by a representative of Electrical Research Products, Inc. 

February 14, 1940. The CBS Broadcasting System Television Station, by 
P. C. Goldmark and John Dyer (at the CBS Studios in the Grand Central Sta- 
tion Building, New York, N. Y.). 

March 13, 1940. Motion Picture Film as Related to Television, by representa- 
tives of the Eastman Kodak Co., Columbia Broadcasting System, National 
Broadcasting System, and Baird Television Corp. 

April 10, 1940. A demonstration and description of the large-size Baird theater 
television screen, by Edward Truefitt, of the Baird Television Corp. 

This list is fairly certain, although it is subject to change in the event of unfore- 
seen circumstances. 

On December 13th, a meeting of the Section was held at the Hotel Pennsylvania, 
New York, at which time Mr. F. E. Carlson of the Lamp Department of the 
General Electric Company presented a paper on "The Characteristics of Vapor 
Light-Sources." The paper was of a tutorial nature, discussing the character- 
istics of vapor sources operating under the combination of variables of pressure 
and of current density. Transmission characteristics of the envelope and other 
pertinent characteristics were included. The practical adaptation of light- 
sources having these characteristics were enumerated as applying particularly to 
projection and printing light sources. 


Results of the election of officers and managers of the Mid- West Section for 
1940 were as follows: 

J. A. DUBRAY, Chairman 
I. JACOBSEN, Sec.-Treas. 
C. H. STONE, Manager 


S. A. Lukes continues on the Board of Managers as Past-Chairman, and also 
O. B. Depue, as Manager, whose term has one more year to run. 

At the November 20th meeting, held in the meeting rooms of The Western So- 
ciety of Engineers, Chicago, the following two papers, orginally presented at the 
October convention of the Society at New York, were re-presented for the benefit 
of the mid- western members : 

"Lenses for Amateur Motion Picture Equipment (16-Mm and 8-Mm)," by 
R. Kingslake of the Eastman Kodak Company, Rochester, N. Y. 

' 'Some Industrial Applications of Current 16-Mm Sound Motion Picture Equip- 
ment," by W. H. Offenhauser and F. H. Hargrove, of The Berndt-Maurer Cor- 
poration, New York, N. Y. 

On December 19th, an additional meeting was held, at which time Mr. M. 
Wenzel of the Wenzel Company, Chicago, presented a paper on "New Develop- 
ments in Theater Motion Picture and Sound Projectors." 

Both meetings were well attended and considerable discussion followed the 


Elections of officers and managers of the Pacific Coast Section are now in prog- 
ress, and the results will be announced in the next issue of the JOURNAL. 

At a meeting held in the RCA Building, Hollywood, on November 20th, two 
papers dealing with television were presented as follows : 

"Discussion of RCA Television Demonstration Unit," by W. C. Turner. 

"Discussion of Television Receiver Installation Problems," by I. Steinberger. 

Following the discussions, an opportunity was presented to inspect the tele- 
vision transmitter-receiver demonstration unit. This is the unit that was dis- 
played at the San Francisco Fair by RCA. 


At a recent meeting of the Admissions Committee at the General Office of the 
Society, the following applicants for membership were admitted to the Associate 


309 E. 23rd St., Piedmont Hotel, 

New York, N. Y. Seattle, Wash. 


OWN, fc. i . 19Q5 Notti ham Way 

Southern Amusement Co.. T 

T , . T Trenton, N. J. 

Lake Charles, La. 


BROWNE, E. A. 81 Green St., 

44 Clement Ave., Brookline, Mass. 

West Roxbury, Mass. HUMBY, W. W. 

BURGER, M. J. 30 Visaat St., 

121-18 109th Ave., Saint John No. 8, 

So. Ozone Park, N. Y. Canada. 




195 Broadway, 

New York, N. Y. 
227 W. 17th St.. 

New York, N. Y. 

Midland Television, Inc., 

Kansas City, Mo. 
24 Nepean Rd., 
Malabal Hill, 

Bombay, India. 
424C Deyo Ave., 

Congress Park, 111. 
15 Derby St., 
Sydney, Australia. 


General Service Studios, Inc., 
6625 Romaine St., 
Hollywood, Calif. 

RYAN, L. F. 

4600 N. Winchester Ave., 
Chicago, 111. 


Columbia Broadcasting System, 
15 Vanderbilt Ave., 
New York, N. Y. 

Vidal 1670, 

Buenos Aires, Argentina. 


52 Vanderbilt Ave., 

New York, N. Y. 

In addition, the following applicants have been admitted by vote of the Board 
of Governors to the Active grade: 


Great Kills, S. I., N. Y. 

National Carbon Co., 
Box 6087, 

Cleveland, Ohio 

1197 Merchandise Mart, 

Chicago, 111. 
General Electric Co., 
601 West 5th St., 
Los Angeles, Calif. 

Loew's, Inc., 
1540 Bread way, 
New York, N. Y. 

22 Stokes Terrace, 
Moorestown, N. J. 

39 Orange Rd., 
Montclair, N. J. 

245 W. 55th St., 
New York, N. Y. 




Volume XXXIV February, 1940 



Report of the Projection Practice Committee 125 

Future Development in the Field of the Projectionist 


Projection Room Planning for Safety E. R. MORIN 134 

The Projectionist's Part in Maintenance and Servicing 

J. R. PRATER 143 

Possible Methods for Encouraging Study by Projectionists .... 


Some Industrial Applications of Current 16-Mm Sound Motion 

Picture Equipment 


A Reel and Tray Developing Machine R. S. LEONARD 168 

Considerations Relating to Warbled Frequency Films 

E. S. SEELEY 177 

Science and the Motion Picture H. ROGER 193 

The Preservation of History in the Crypt of Civilization 

T. K. PETERS 206 

New Motion Picture Apparatus 

A New High-Quality Sound System 

Simplex Double-Film Attachment 


A Non-Intermittent Motion Picture Projector 


Current Literature 232 

Book Review 234 

1940 Spring Convention at Atlantic City, N. J., April 22nd-25th 235 

Society Announcements 238 





Board of Editors 
J. I. CRABTRBE, Chairman 




Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscription or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., East on, Pa., or Hotel Pennsylvania, New York, N. Y. 
Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 

West Coast Office, Suite 226, Equitable Bldg., Hollywood, Calif. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1940, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is not 
responsible for statements made by authors. 


* President: E. A. WILLIPORD, 30 East 42nd St., New York, N. Y. 

* Past-President: S. K. WOLF, RKO Building, New York, N. Y. 

* Executive Vice-f 'resident: N. LEVINSON, Burbank, Calif. 

** Engineering Vice-President: D. E. HYNDMAN, 350 Madison Ave., New York, 
N. Y. 

* Editorial Vice-President: J. I. CRABTREE, Kodak Park, Rochester, N. Y. 

** Financial Vice-President: A. S. DICKINSON, 28 W. 44th St., New York, N. Y. 

* Convention Vice-President: W. C. KUNZMANN, Box 6087, Cleveland, Ohio. 

* Secretary: J. FRANK, JR., 356 W. 44th St., New York, N. Y. 

* Treasurer: R. O. STROCK, 35-11 35th St., Astoria, Long Island, N. Y. 


* M. C. BATSEL, Front and Market Sts., Camden, N. J. 

* J. A. DUBRAY, 1801 Larchmont Ave., Chicago, 111. 

** A. N. GOLDSMITH, 444 Madison Ave., New York, N. Y. 
** H. GRIFFIN, 90 Gold St., New York, N. Y. 

* P. J. LARSEN, 29 S. Munn Ave., East Orange, N. J. 

* L. L. RYDER, 5451 Marathon St., Hollywood, Calif. 

** A. C. HARDY, Massachusetts Institute of Technology, Cambridge, Mass. 

* H. G. TASKER, 14065 Valley Vista Blvd., Van Nuys, Calif. 

* Term expires December 31, 1940. 

** Term expires December 31. 1941. 


Summary. A brief account of the work of the Committee during the past year, 
in which is traced the evolution of the SMPE Projection Room plans and their general 
adoption by the industry. Work is being initiated by the Sub- Committee on Theater 
Structures in a study of the disparity that exists among various state and municipal 
motion picture regulations. The subject of heating projection rooms is also reported 
upon by the Sub-Committee on Fire Hazards. Work has been begun by the Sub- 
Committee on the Power Survey in determining the average or representative operating 
conditions, with particular respect to the power consumed, in theaters of various seating 
capacities and equipped with various types of projection apparatus. 

During the past year, the Committee has undertaken and success- 
fully brought to fruition several major projects of outstanding im- 
portance to the technologic groups of the motion picture industry. 
Moreover, the Committee is gratified to note that its recommenda- 
tions to the industry have been widely accepted and put into general 
practice. Indeed the principal purpose of the Committee is to in- 
vestigate any problem of projection that may be facing the industry, 
to determine the existing circumstances, and to recommend any 
alterations in these circumstances that may be dictated by the needs 
of good projection. 

Perhaps one of the most important of the projects that has been 
undertaken by the Committee is the revision of the earlier projection 
room plans. The first set of projection room plans was drawn up by 
the Committee in 1930, and since that time the many advances in the 
art of projection and in the projection equipment have made it 
necessary to revise the plans, the latest edition being that published 
in the November, 1938, issue of the JOURNAL. Naturally, the estab- 
lishment of a set of plans aiming toward a high level of projection, 
operation, maintenance, and safety would lead toward a thorough in- 
vestigation of projection room design from the point of view of fire 
prevention and control. These plans and survey have been published 
in The Architecture Record, thus bringing them to the attention of 
architects of the country and so contributing toward improvement 
in theater design. In the same issue of the JOURNAL was published a 

* Presented at the 1939 Fall Meeting at New York, N. Y. ; received October 15, 



comprehensive revision of the "Regulations for Handling Nitro- 
cellulose Motion Picture Film" of the National Fire Protection 
Association, the original edition of which was brought out in 1931. 
Obviously, in view of the marked advances in the art since 1931 
these regulations required thorough. revision. 

After study of the proposed revision, by the Committee on Hazard- 
ous Chemicals and Explosives of the NFPA, the revision was voted 
upon at the May, 1939, meeting of the NFPA at Chicago and was 
adopted practically without change. A few additional points not 
covered in the present revision still remain, but these features will be 
taken under advisement by the Projection Practice Committee during 
the coming season. The NFPA has recently issued a revised booklet 
of regulations, which may be obtained upon request to them. 

The revision of the regulations was based in large part upon the 
great improvement that has occurred during recent years in the con- 
struction of motion picture projection rooms and this improvement, 
the Committee feels, is largely due to the Projection Room Plans 
originally issued in 1930 and periodically brought up to date. One of 
the most interesting results of this work was a recent ruling in the 
State of Connecticut with regard to the use of sprinklers in motion 
picture projection rooms. The SMPE projection room plans form 
the basis of the Connecticut State Regulations and, in turn, the New 
England Fire Insurance Rating Association has ruled that in motion 
picture projection rooms, constructed and protected in accordance 
with the requirements of the State of Connecticut, automatic sprin- 
kler protection may be omitted from the projection room. It is further 
recommended by the New England Fire Insurance Rating Associa- 
tion that the soda-and-acid chemical extinguishers be placed im- 
mediately outside the projection room rather than inside. 

It has long been the opinion of the Projection Practice Committee 
and the Sub-Committee on Fire Hazards that automatic sprinklers 
might well be omitted from projection rooms in view of the high 
speed with which film burns and the great damage done to the sound 
and projection equipment when the automatic sprinkler valves are 
released. A reel of film set on fire will often burn itself out before the 
sprinklers have had a chance to operate; then afterward the heat of 
the room may release the sprinklers. In projection rooms designed in 
accordance with the SMPE Projection Room Plans and the require- 
ments of the State of Connecticut, the chances of fire in such projec- 
tion rooms are very much lessened and, in addition, the transmission 


of the fire beyond the confines of the projection room so constructed is 
practically negligible. It is on these bases that the State of Con- 
necticut and the Committee recommend that sprinklers be omitted, 
and it is a source of satisfaction to the Committee to be able to trace 
this evolution of practice back to its original SMPE Projection 
Room Plans. 


One of the most troublesome features of projection room design, 
construction, and installation of equipment is the great disparity 
that exists among various state and municipal motion picture regula- 
tions. The extent to which these various codes differ and conflict 
has been very forcibly shown by the Board of Labor statistics of the 
United States Department of Labor in a publication in the Monthly 
Labor Review of January, 1938, which dealt specifically with "Safety 
Standards for Motion Picture Machine Operators.'.' It is not neces- 
sary at this point to go into these differences, as the original publica- 
tion in the Monthly Labor Review may easily be consulted. However, 
it is well known that such disagreement and conflict among the codes 
of the states and municipalities exist not only with respect to pro- 
jection rooms but also in relation to theater structure and auditorium 
conditions. The Sub-Committee on Theater Structures has been 
engaged in studying these problems, the latter subject, namely, 
auditorium conditions, including both lighting and acoustic treat- 
ments, as well as other factors related to the viewing and hearing of 
motion picture productions. 

The first step in the work is to obtain copies of the codes of all the 
states and important municipalities. When these codes have been re- 
ceived they will be compared and analyzed in the hope that ultimately 
a representative code might be drawn up for submission to the various 
states and municipalities, with the view of fostering uniformity in 
their regulations. This is admittedly an ambitious program. How- 
ever, at least one state, namely, Connecticut, is already working very 
closely with the Committee, and the influence of the Committee's 
work is being felt throughout the country by virtue of its recent rec- 
ommendations toward revision of the NFPA regulations, so it is 
hoped that other states may be induced to join in the work. 

Following naturally upon the projects described above comes the 
problem of heating the projection room. Information has come to the 
Sub-Committee that hazardous practices are found in the industry 


with regard to heating projection rooms. Instances have been cited 
wherein open-flame gas heaters have been used. 

The subject of heating projection rooms has already been covered 
by the Projection Practice Committee in the Projection Room Plans, 
wherein it is stated that "proper provision shall be made for heating 
projection rooms. The same facilities used for heating the theater 
should be extended to the projection room." 

The regulations of the NFPA for handling nitrocellulose motion 
picture film state that "artificial heating in any building or room 
other than a vault in which motion picture film is used, handled or 
stored, shall be restricted to steam. . . .Heat generating apparatus 
shall be in a separate room .... Ordinary hot-air furnaces are pro- 
hibited. Gas, oil, and electric heaters are prohibited in rooms where 
film is handled or stored." 

The problem of heating projection rooms may be approached from 
two points of view: the ideal or the immediately practical. Writing 
regulations into the SMPE Projection Room Plans, which aim toward 
the ideal would lead to connecting the projection room to a central 
heating plant. In instances where this is not convenient, economic, 
or practicable provision for local and separate heating of some pro- 
jection rooms may have to be made. Even when a central heating 
plant is used, the projection room may require heating before the 
show to assure fluidity of the oil and grease in the projector, and to 
make the projection room temperature bearable to the projectionist, 
especially when the projection room is located on an outside wall of 
the building in an exposed situation. 

This Committee definitely does not recommend the use of gas or 
oil for any individual heating unit located inside the projection room. 
This does not mean that gas may not be used for central heating 
plants using steam, hot water, or hot air, or for any auxiliary heating 
system in which the heating element is remote from the projection 
room. This report is not concerned with heating the theater audi- 

For individual units in the projection room, electric heating is the 
only alternative. Although the NFPA regulations prohibit the use of 
electric heaters, this regulation was written in 1931, at which time 
no satisfactory electric heaters were available. Under no circum- 
stances does the Committee recognize as suitable for projection room 
use the old conventional type of electric heater, or any other kind of 
electric heater having exposed heating elements. For any type of 


individual electric heater that may be used, the heating element 
should be completely enclosed and this entire heating assembly and 
any blower or fan should be enclosed within a protective housing. 
The temperature of the outside surfaces of this housing should never 
rise above 115F. This may require, in the design of the unit, pro- 
vision for automatically shutting down the unit in case of a rise of 
temperature above this value or failure of the blower. The protec- 
tive housing shall be so constructed that no pieces of scrap film or the 
like may be able to enter through any openings or louvers provided 
to permit the exit of heated air or the entrance of cool air. The top 
of the casing should be made to slope or be rounded so that it can not 
be used as a shelf. 

The heating unit (and blower) must conform in electrical design, 
wiring, installation, and circuit connections, to the requirements of 
the National Electrical Code. 

The Sub-Committee on Fire Hazards has investigated the portable 
heating equipment available on the market and has found at least 
one portable electric heater designed and constructed in such fashion 
as to fulfill substantially all these requirements. The heater is 
available in several sizes, from 1 kw up, and has two degrees of heat- 
ing that is, provision is made for delivering heated air at a tempera- 
ture of either 45 or 60 F above the ambient room temperature. 


Other Sub-Committees of the Projection Practice Committee are 
engaged in projects not yet carried to the point where definite reports 
may be made. The Committee on Projector Output and Screen 
Illumination has done considerable work on methods and devices for 
measuring the light incident upon projection screens and reflected 
therefrom. The measurement of the incident light does not present 
any problem, in view of the fact that there is available on the market 
a simple, inexpensive light-meter with a built-in color-filter for 
simulating visual response to color. 

The problem of measuring reflected light in a convenient, inexpen- 
sive, and simple way with equipment not requiring highly trained 
specialists, has not yet been solved. The Committee is investigating 
the problem further and hopes to be able to report more definitely 
at the next Convention. 

The work of the Sub-Committee on the Power Survey, described in 
the previous report of the Committee, published in the July, 1939, 


issue of the JOURNAL, is proceeding satisfactorily, and the Sub-Com- 
mittee reports that approximately 600 of the questionnaires have 
been returned. The purpose of this study is to break down into its 
components the total electric power used in motion picture theaters, 
so as to determine the average or representative operating conditions 
in theaters of various seating capacities and equipped with various 
types of projection apparatus. The work has about reached the 
stage where analysis of the data may be undertaken and it is hoped 
that a comprehensive report may be available by the time of the next 

The Sub-Committee on Screen Border Illumination is investigating 
several methods of screen border lighting, some of which has received 
practical demonstrations in motion picture theaters, and one of 
which is in current use. Material on this subject for a possible 
future report is in the course of preparation. 

H. RUBIN, Chairman 










DR. GOLDSMITH : You will notice that the Projection Practice Committee has 
conducted extensive surveys to secure data of various sorts relative to theaters 
located throughout the United States, to assemble and correlate these data, and to 
study and analyze them in order that any exhibitor will be in a position to deter- 
mine whether his theater corresponds to average practice, better than average, or 
worse than average theatrical practice. It is hoped that such information will 
be of help not only to the exhibitors in determining the technical or economic 
correctness of their present practice, but also to the architects of future theaters; 
and that it may in time serve as at least a partial guide in the design of theaters 
of more economic, practical, and attractive construction. 

I may add that there have appeared before the Committee from time to time, 
and at its invitation, individuals who have presented on behalf of their companies 
or themselves pieces of equipment (such as heaters, measuring instruments, 
photometers, and the like), in order that the Committee might consider and study 
these. The Committee, of course, takes no part in any commercial issues but it 
does technically analyze devices and methods and consider their practicability. 


Summary. The highly diversified activities required for the production of a 
motion picture find their effective culmination in the work of the theater projectionist. 
The unusually concentrated value embodied in the reels of film corresponding to a 
feature picture can be brought to the theater audience and made the basis for com- 
mercial returns only through the activities of the projectionist. 

Nevertheless the public is little aware of what goes on in the projection room. 

The projectionist is in part compensated by the likely stability of his activities. 
His present position in the theater is important. Future developments in the motion 
picture field, such as three-dimensional sound, wider use of color, and the like, will 
make his work even more important. The possible inclusion of television projection 
in theater programs will require his mastery of the new field which is sufficiently 
similar to his present activities in its broad outline to enable its handling by the theater 

It has seemed well in the past that the Society of Motion Picture 
Engineers should wisely devote close attention of an appropriate 
technical committee to the problems of motion picture projection. 
This task has in fact been ably accomplished by the Projection 
Practice Committee which has effectively considered the numerous 
devices, arrangements, and procedures which are involved in theater 
projection. It is appropriate at this point to comment on the activi- 
ties of the projectionist himself, since good equipment alone is not a 
complete answer to all projection questions. 

The position of the projectionist in the motion picture field is 
rather a peculiar one. He is taken for granted by practically every- 
body and it is calmly assumed that his difficult job will be done 
and done well. The producer, the distributor, the exhibitor, and the 
audience alike take it for granted that the picture and sound will be 
delivered satisfactorily and that "the show will go on." In a way, 
this is a high compliment. If a job is usually done so thoroughly 

* Presented at the 1939 Fall Meeting at New York, N. Y. ; received October 
10, 1939. 

** Past-President, SMPE; Past-President, IRE; Consulting Engineer, 
New York, N. Y. 


132 A. N. GOLDSMITH (j. s. M. P. E. 

and well that it becomes almost a routine matter and that every- 
body expects it to be satisfactory, it is a tribute to the care and con- 
sistent effort qf the projectionist. 

Yet I think that this attitude fails to appreciate the importance of 
the projectionist in the motion picture set-up. We hear all about the 
glamorous stars in Hollywood, but every bit of this glamor has to 
pass through less than a square inch of film gate to reach the audience. 
We are impressed by the elaborate stage sets and studio equipment 
on the West Coast, but it is the projectionist who delivers the results 
through the projection room port. Millions of dollars and years of 
time of many people may be spent on a single feature film. Authors, 
writers, actors, directors, producers, cameramen, sound recordists, 
electricians, and a host of other studio personnel, as well as the 
laboratory and exchange workers may be required to deliver the 
feature film to the theater. And then, the projectionist must "de- 
liver the goods" or take the consequences without acceptance of 
excuses. It is curious that most other workers in the industry occa- 
sionally receive public notice and praise. There are few occasions on 
which the projectionist receives public recognition or acclaim. Yet 
it would be well if the public understood that they meet the wide- 
spread skill of the projectionist when they see good pictures, even 
though the projectionist quietly does his job back of the scenes. 

A modern projection room from which black-and-white or color 
pictures and sound are projected is a place full of complicated equip- 
ment which requires skillful handling. The alert and capable pro- 
jectionist of today has to know a lot more about pictures and 
sound than his predecessor of twenty-five years ago. Further- 
more, the projectionist of tomorrow will have to know still more. 
Looming on the horizon are increases in the use of color pictures 
which mean, in turn, a brighter and whiter screen with careful control 
of illumination intensity and color. New types of projection, of 
screens, and of theater design are all in the offing. Three-dimensional 
sound where the sound of the speaker appears to follow him around 
the screen is one of our prospects. And, most startling of all, tele- 
vison projection is closer than "around the corner." 

Speaking of television, it is interesting to know that in England 
television pictures as large as 15 X 20 feet are projected on the 
theater screen. While these pictures do not have either the full 
brightness or detail of present motion pictures of the same size, yet 
they have been good enough to fill large theaters repeatedly and to 


induce the theater chains to order the installation of substantial 
numbers of such television theater equipments of various types. 

There are several fundamentally different sorts of television pro- 
jectors for theaters, and no one is sure just what will constitute 
"standard" theater equipment in that field 5 or 10 years from now. 
However, it is reasonably certain that enterprising showmen will 
find timely and entertaining material suitable for theater presentation 
and that mixed film and television programs will gradually be ac- 
cepted in the theater field. 

I regard this as one of the finest and most encouraging prospects 
which the projectionist faces. The optical principles governing tele- 
vision projection may differ in detail from those used in film projec- 
tion, but broadly they are quite similar. The sound reproduction in 
the television program is of course carried out by amplifiers and loud 
speakers as at present. The enterprising and up-to-date projection- 
ist can master television projection as readily as he did film projection 
and can make himself just as invaluable in the theaters of the future 
as he is in the theaters of today. The projectionists and their organi- 
zations might well study this new field and keep up to date on it so 
that, as it finds a place in the theater, they may be prepared to assume 
an important position in the television field as well. 

There are some lines of work where a man today might feel puzzled 
or worried as to his future. He might wonder whether there was going 
to be a demand for the commodity or services which he produces. 
Or he might doubt whether his field would hold its own against some 
new competitor. The projectionist of today is peculiarly fortunate. 
If he is energetic and determined in the future he will be free from the 
dangers I have just mentioned. His field is an expanding field, with 
new opportunities and obligations. He and the engineers should 
draw closer to each other in an association of substantial help to each 
group. Thus the projectionist will, in all likelihood hold the same 
key positions in the theaters of the future as he does in those of today. 

E. R. MORIN** 

Summary. A brief account of the work of the State of Connecticut with respect to 
the abatement of fire and other hazards incident to the presentation of motion pictures. 
A model projection room, the design of which has been based upon the experience of the 
State of Connecticut in the motion picture field, is described and illustrated. 

The purpose of this paper is to relate the experience of the Depart- 
ment of State Police of the State of Connecticut with respect to the 
abatement of fire and panic hazards incident to the presentation of 
motion pictures. 

Progress and success in the control of hazards can result only from 
close observation and the experience gained thereby, and it has been 
found that the closer the observation and the longer the experience, 
the better the chances are for improvement. 

By virtue of legislation passed in 1909 the Department was given 
control of the subject and at that time both the industry and the 
equipment were in a primitive stage. The theaters were mostly su- 
perannuated "opera houses" and so-called moving picture theaters 
of the shooting gallery type set up hi vacant stores. Film fires were 
expected and these expectations were promptly realized, all of which 
was food for thought as well as an excuse for high insurance rates. 

Conservative men were assigned to the enforcement of the law, 
and by diligent study of the problems and firm administration of the 
rules, they gained and maintained the respect of both the theater 
owners, underwriters, architects, and contractors. 

Material savings were made but in general nothing of great impor- 
tance occurred until 1921 when there was a serious catastrophe caused 
by a stage fire in a combination vaudeville and picture house. This 
fire did not originate from picture equipment but it stirred up the 
population of the state relative to safety in places of public assembly 
and it gave the state police the opportunity of making and enforcing 

* Presented at the 1939 Fall Meeting at New York, N. Y.; received October 
6, 1939. 

** Department of State Police, Hartford, Conn. 



stronger and better regulations covering all places where motion 
pictures were exhibited. Since that time there has not been a fatality 
or a large fire loss that could be charged to the use of film, and since 
that time state policemen assigned to inspection of theaters have 

FIG. 1. 

Complete model of projection room (width about two feet) viewed 
from the rear. 

been relieved from all other duties and been required to make inten- 
sive study of both building and projection. 

The Department's next great opportunity was ten years ago when 
everybody in the picture business went in for sound or went out of 
business. This revolution in projection necessitated larger projec- 



LJ. S. M. P. E. 

tion rooms and the Department sold the idea of building not only 
larger but better projection rooms. 

When a new theater is to be built or an old one reconstructed we 
supply the architect with photographic copies of plans and illustrations 
of how a model projection room should be designed and equipped. 
In connection with this the Department has recently installed a com- 
plete projection room with a small auditorium in our Administration 

The Department has maintained membership in the Society of 
Motion Picture Engineers, the National Fire Protection Association, 

FIG. 2. Fiont view of model, showing ports. 

the New England Building Inspectors Association, the New England 
Fire Chiefs' Association, and has worked in close cooperation with 
the underwriting organizations who pass on special hazards. In- 
surance rates are or ought to be merely a reflection of the hazards in- 
volved, and the abatement of hazards either does or should mean a 
reduction in insurance premiums. We expect to make the exhibition 
of motion pictures in Connecticut so safe that theaters will be taken 
out of the special hazard class. 

Much has been learned about buildings and projection through 
our association with organizations, particularly the Society of Motion 


Picture Engineers which the Department joined in 1936, and its 
representative has been an active member of the Projection Practice 
Committee. The Department has found this engineering body very 
helpful incoordinating with them in several of their technical prob- 
lems. The real school for learning is the careful observation incident 
to our day-to-day examination of theaters in action. 

Out of our experience in these matters it was determined that the 
way to cope with burning film was to provide absolutely fireproof pro- 
jection rooms built of masonry, so that in the event a careless pro- 
jectionist, acting in error, should have a few thousand feet of film ex- 
posed, it might be burned without loss or panic outside the projection 
room. Our safety devices are automatic in action, and about all the 
projectionist has to do is to get out of the projection room and see 
that the door is closed. 

Examining these activities in detail I bring to your attention the 
following : 

About ten years ago we experienced an explosion in one of our 
projection rooms. The result of our investigation proved one 
thing: that we did not have the proper ventilation. We found that 
the 10-inch indirect-type flue did not give us a natural circulation of 
air due to stack resistance. Furthermore, the conventional fan used 
in that period was not sufficient to exhaust the smoke and gases 
which were generated from burning film. We learned also that a 
vent going through the rear or side wall of a projection room was 
subject to back-drafts, depending upon weather conditions. We 
then conducted experiments with vertical flues, eliminating elbows, 
increasing the diameter from 10 to 18 inches, and installing a bucket- 
blade fan instead of a flat-blade fan. A further survey disclosed that 
the numerous types of hoods used on these flues offered considerable 
resistance which cut down the desired natural circulation of air. We 
then investigated several types of hoods until we found the type which 
offered the least amount of resistance. Over a period of ten years 
this combination has proved very satisfactory. 

Our next step led us to the conclusion that, regardless of unproved 
ventilation, less damage would accrue if the booths were larger. Mov- 
ing hi that direction, we specified projection rooms with a minimum 
depth of 12 feet and a width of 16 feet and a height of not less than 
8 feet. One of our primary requirements with respect to size is that 
no protruding object shall be within 30 inches of the right or rear of 
any projector. 



(J. S. M. P. E. 

We also felt that the old conventional type motion picture booth, 
consisting of asbestos board and angle-iron, did not afford the neces- 
sary fire-resisting segregation from the auditorium. Furthermore, 
this type of booth, with the advent of sound, permitted the noises of 
the projection room to be transmitted to the auditorium. 

Considering the above facts, we decided on a masonry type of 
projection room, but in so doing we were confronted with another 
problem. We found that quite a number of the older theaters were 


2* HIM. WOOL. 











FIG. 3. Cross-section of projection room. 

not structurally designed to receive the added weight. We then had 
to design a projection room eliminating as much weight as possible 
but without jeopardizing any of the safety factors. This was accom- 
plished by erecting a masonry or steel foundation with a steel frame, 
4-inch reinforced concrete floor, 4-inch gypsum-block walls and ceil- 
ing, with hard plaster inside and a standard or acoustic plaster out- 
side, using a fire-door equipped with a door check but no latch, and 
permitting only a dead lock. The reasons for this latter feature are, 
first, in the event of a fire, if the pressure in the projection room should 

Feb., 1940] 



reach the explosion point, the door being the weakest point, should 
open and close automatically; and, second, that should the hands and 
face of the projectionist be burned he should be able quickly to leave 
the projection room without having to fumble about for the door 

In the past it was customary to use a ladder through a trapdoor 
in the floor or a ladder along the side of the wall to reach the projec- 
tion room. This practice has been discontinued by permitting only 
a fire-resistant stairway with the minimum treads of 9*/2 inches and 
maximum risers of 8 inches. 



FIG. 4. Front elevation of projection booth. 

In the past the port shutters were fastened with string and fuse 
links along the ceiling. The objection to this was that a considerable 
flame was necessary before the shutters would operate. The manual 
release was either at the rear of the projection room or close to the 
door. We found that this condition could be improved upon. We 
proposed speeding up the closing of the shutters by erecting a rod over 
the shutters with pins in the rod to which the shutters are attached by 
means of a cord and rings. Also opposite each projector there is 
provided a protruding arm with an eye in each end through which an 
endless cord is run from one projector to the other and attached to 
fuse links which, in turn, are attached to the rear part of the upper 
magazine near the door latch. By so doing the fuse link is placed as 
nearly as possible to the point where the fire would probably originate. 



(J. S. M. P. E. 

Also it should be recalled that the projectionist is at hand to shut his 
machines down and he can release the fuse link and drop the shutters 
without going to another station. 

In the event of a fire the projectionist is supposed to shut down 
his machines, close the port shutters, turn on the house lights, and 
see that the projection room exhaust fan is in operation. Through 
past experience we have discovered that in some cases the projec- 
tionist did not turn on the house lights or the projection room ex- 

! I f 9" 3-3" i l'9"i 

ffi iijo--- 3PSfc 
* . j . . 

'l-VI'lil. ' "' 1 j 1 1 

-1 ti-' * | V > 

; 18* 





15" VENTS *( 



/' " s 
i i 






4"x 1 5 " VENTS 4"x 1 5" VENTS 

3 x i i a 

FIG. 5. Reflected ceiling plan of projection room. 

haust fan. To overcome this difficulty the Department designed a 
switch, the details of which were published in the May, 1939, issue 
of the Society's JOURNAL. 

We also encountered difficulty with the conventional type of film 
cabinets. With the cooperation of an equipment manufacturer these 
cabinets have been unproved by inserting a heat-resisting material 
which completely fills the air-space between the compartments and 
the cover. This prevented the external heat from burning film in the 
vicinity or adjacent to the cabinet and from igniting films in the 


In recent years we have encountered a new difficulty with the 
installation of refrigeration and air-conditioning equipment. First, 
in the event of a fire and the failure of the port shutters to close, and 
the breaking of the glass covering them, the ports would become 
exhausts due to the auditorium ventilation. This possible condi- 
tion was overcome by connecting the auditorium ventilation to the 
emergency switch mentioned above. Second, the conventional 
type of intake ventilation near the floor became an exhaust instead of 
an intake. This condition is now under survey, and these intakes are 
being taken from the outside. We have received some objections 
from the projectionists to this particular method ; in particular, they 
have objected because the projection room would be too cool in the 
winter. We are at the present experimenting to find a solution for 
this condition. 

The Department, realizing that this is a highly technical subject, 
has purchased numerous instruments to enable this survey to be 
conducted in the proper manner. Projection room ventilation is a 
large subject in itself, and as the Department survey is not complete 
I shall not here attempt to discuss it further. 

Our records show that the majority of our film fires have been 
caused by faulty patches. Our method of overcoming this difficulty 
is by insisting that all projection rooms be equipped with a mechanical 
film splicer. 

The Department has been successful in obtaining from the New 
England Insurance Rating Bureau the following exemption, a section 
of which is here quoted : "That based upon the motion picture booths 
constructed and protected in accordance with your requirements we 
are agreed to omitting automatic sprinkler protection from inside 
the booth. Furthermore, we desire to go on record as recommending 
that the soda and acid chemical extinguishers be placed immediately 
outside of the booth rather than inside." 

In conclusion, three important results have been accomplished: 
first, greatest safety for all concerned; second, guarding the health 
and convenience of the projectionist; and, third, better working con- 
ditions, which always lead to a better show for the patrons. 


MR. EDWARDS : This is one of the most practical reports we have had in a long 
time; an example of a State cooperating with the Society (and I think it is the 
only State so far that does so), employing the same degree of investigation as we 
do here. 

142 E. R. MORIN 


MR. RICHARDSON: In many projection rooms the fuses are placed 6 or 8 feet 
from the possible source of fire. I have proposed that a film link be so situated 
that any fire occurring in the projector mechanism or at the rewinder would al- 
most instantly strike the link and sever it. Do you think the metallic fuse serves 
the purpose equally well? 

MR. MORIN: I think it does. We do not depend wholly upon the automatic 
fuse. By placing the control right next to the projectionist, it is very easy for 
him to trip it and throw off the motor and lamp switches in one operation. Years 
ago I used a film, but the film becomes brittle with age, and if it is not changed 
occasionally the shutter may drop when you do not want it to drop. 

MR. PATENT: The projectionist may not be at the right spot at the moment 
of the fire, to operate the controls instantly. I have in mind an electrical release, 
which could be operated from any point desired. 

MR. MORIN: We do not wish to make things more complicated, but rather to 
make them as simple and practicable as possible. The State of Connecticut re- 
quires that the projectionist stay close to his projector, so it should not be neces- 
sary to have additional gadgets. 


Summary. It is the duty of the projectionist to see that all projection equipment 
is kept in condition to give excellent service dependably and efficiently. It is impossible 
to accomplish these results by depending upon memory alone. The projectionist 
must establish and keep written records of all necessary maintenance data. He 
must follow a written schedule in making inspections and in doing maintenance 
work. He must establish a reliable system for checking and ordering supplies and 
spare parts at regular intervals. 

The projectionist should do as much of actual service work as his knowledge, ability, 
tools, and available test equipment will permit. A t least nine-tenths of trouble shooting 
should be done before any trouble exists. He should obtain detailed drawings of in- 
ternal and installation wiring of all electrical equipment, besides identifying the 
points at which tests may be made. He should prepare a written outline of all tests 
that could be made if various troubles existed. Then he should actually make all 
possible tests in advance, wherever possible, without causing damage, by deliberately 
creating the trouble and then correcting it. He should immediately record the exact 
results of each test in the written outline. In this way, simple test s may serve as well as 
or better than elaborate ones. 

The professional service engineer with special test equipment is a necessity to the 
finer and more difficult parts of modern servicing, but the projectionist who makes the 
best of what resources he has can also do a very valuable part of the job. 

Motion picture projection equipment requires a considerable amount 
of maintenance and servicing in order that it may function at all. 
However, the purpose of good maintenance reaches far beyond this 
minimum. When patrons pay admission to a theater the show must 
go on. If they are to return again and again the show must be not 
only good in itself, but also it must be well presented. Therefore 
dependability of equipment and excellence of results they produce 
are first considerations. At the same time it must be remembered 
that the theater is a business that must support itself and show profit 
in order to succeed. Therefore its equipment must be purchased, in- 
stalled, maintained, and operated without unnecessary expense. 

* Presented at the 1939 Fall Meeting at New York N. Y. ; received August 
30, 1939. 

** Congress Theater, Palouse, Wash. 


144 J. R. PRATER U. s. M. p. E. 

First-cost and installation expense must be as low as possible consis- 
tent with good results. After that, every part of every equipment 
must be made to perform its duty faithfully and well and for the long- 
est possible space of time, though in every instance necessary adjust- 
ments or replacements must be made before serious trouble or com- 
plete failure occurs. These are often conflicting requirements, and 
the following suggestions will serve a very useful purpose if they 
merely assist in arriving at a happy medium : 

All real projectionists would rather do good work than bad, but 
many fail to get started in the right way. Equipment maintenance is 
a splendid example, because establishing a system which will make it 
possible for them to do a better job of taking care of all the projection 
equipment entails a bit of extra work. However, once such a main- 
tenance system is started and adjusted to fit the needs of the individ- 
ual projection room, the projectionist's regular duties can be per- 
formed so much better and with so much less effort that very soon an 
actual saving in work will result. 

No projectionist, regardless of his ability, can keep in his head all 
the data necessary to good maintenance of all the various equip- 
ments under his care. Neither can any one man depend upon being 
immediately available at all times when such data may be needed. 
It is therefore absolutely necessary to good maintenance that com- 
plete written records be kept in the projection room. 

A few theaters, mostly in the de luxe class, already keep some sort 
of written records. The average small house the very ones that 
need an efficient system the most because equipment is less reliable 
and operating budgets are strictly limited usually has no records 
at all. 

Because every theater has different kinds and amounts of equip- 
ment the most practicable way of starting such records is in loose-leaf 
form. With reasonably careful handling, good loose-leaf books can 
be made to last for years, but if desired the records can be transferred 
to permanently bound books as soon as the system has been properly 
established and adjusted to fit the individual projection room. In 
any case it is very important to provide a convenient place to keep 
the record books, and to see that they are always kept there when not 
actually in use. 

Before we can keep records of equipment units and parts we must 
be able to identify each part throughout its entire useful life. Where 
there are two or more duplicate units such as projectors, rectifiers, 


amplifying channels, etc., each should be permanently numbered. 
In addition, individual parts or assemblies including all spares that 
may be transferred from one unit to another during their useful lives, 
such as, for example, intermittent units, photocells, amplifier tubes, 
and so on, should be individually numbered. 

Now, taking each large unit separately, write in the record book all 
available information that may later prove valuable for reference. 
Use the first sheet for general data concerning the unit as a whole. 
Start a new sheet for each part or small assembly, leaving plenty of 
blank space for future entries in the case of parts subject to wear or 
deterioration. The loose-leaf binder will permit additional sheets 
to be added for unexpected future needs. 

As an example, under "Projector No. 1" place the date installed, 
total cost, names of dealer and installation engineer, the nature and 
cost of any known replacements or service since installation, date of 
present inspection, and general working condition. On the next page 
list the upper magazine, giving the date of inspection, present condi- 
tion of individual parts such as door, hinges, and latch, observation 
windows, spindle, key, bearing, friction clutch, valve, and all valve 
rollers. Another page may be devoted to the upper feed-sprocket, 
its stripper, shaft, bearing, and gear. The entire projector is thus 
divided into small assemblies or closely associated groups of parts. 
The same procedure is followed for Projector No. 2, and so on for a 
all other equipment. This will require some time, but when it is done, 
we have a foundation of the entire maintenance system. 

Keeping all projection equipment in condition to give excellent 
and reliable service requires that every part shall receive attention as 
often as required. This amounts to a very considerable task, so it 
is also important that unnecessary attention be avoided. Since the 
expected lives of various parts and units may range from a few min- 
utes to several years, some sort of regular schedule must be followed 
to insure proper attention. In this case, again, no projectionist can 
depend on memory alone. A written reminder is necessary, at least 
for those items requiring attention only occasionally. It is easy 
enough to remember to inspect carbon trims after each reel ; to clean 
or oil those places requiring the same attention every day. How- 
ever, it is not so easy to remember when amplifier tubes should be 
checked, generator bearings oiled, switch contacts cleaned, the screen 
dusted, and such other important but less frequent jobs. To insure 
attention to these jobs before trouble occurs and the results as viewed 

146 J. R. PRATER [J. S. M. p. E. 

and heard by the public are injured, the projectionist must follow a 
written schedule. At first such a schedule can be only a guess at the 
proper intervals between inspections and attention but corrections 
can be made whenever they are found necessary until a reliable guide 
is established. It will usually prove better to distribute the less fre- 
quent jobs so that they do not create congestion of work at any one 
time, always following the same order, so that every job is repeated at 
the proper interval. 

And now for supplies and spare parts. The first requirement is 
to have a definite and proper place to keep them. Unless there is 
already such a place, provide it at once, calling on the management 
for a cabinet, shelves, or whatever is really necessary. Be reasonable 
about material and do not be reluctant to do some of the work your- 
self, but do not shirk labor by doing an inferior job. Next, it is 
necessary to have some reliable means of keeping a sufficient stock of 
spare parts on hand to care for all ordinary needs, without necessity 
for ordering each item separately at the last minute. A good plan 
is to keep a "Want List." Whenever routine inspection reveals 
that certain parts will need replacing within the next month or two 
record them on the "Want List." Make a written estimate of what 
supplies and spare parts should be regularly kept on hand. For each 
item estimate the quantity you may reasonably expect to need during 
at least one month, plus ample time to obtain replacements. At the 
beginning of each new month, compare this estimate with the actual 
stock on hand, and add to the "Want List" everything necessary for 
another month's run. 

Immediately turn this list in to the manager. It is a splendid 
policy to have him sign a duplicate copy to be kept in the projection 
room. This copy may well be made on a loose-leaf sheet that can be 
immediately put into the record book binder. This protects both the 
projection room and the projectionist. It enables the manager to 
order everything necessary for a normal month's operation at one 
tune. This policy will not often require any added investment for 
the theater, since practically all business accounts may be paid 
the month following the order. 

With a complete system of this kind once established and adjusted 
to fit the individual projection room, the projectionist's mind is re- 
lieved of remembering what has been done, what needs to be done at 
once, and what can safely be left until tomorrow or next week. He 
can have confidence in his equipment; he can take pride in how well 


and how efficiently it performs. His mind is free to concentrate on 
the actual performance of his duties. 

For some reason, projectionists too often regard service work of 
any kind as entirely out of their line. No doubt lack of knowledge, 
ability, and tools, or the plain desire to avoid are often factors. The 
finer points of modern sound servicing really have outgrown the re- 
sources of the projection room and the projectionist. As one very 
able sound engineer has put it, "Efficient, fast-moving sound ser- 
vice required by conditions today calls for more than a competent 
inspector with a voltmeter and an educated 'nose for trouble.' ' 
True enough, the entire job can no longer be done by the projection- 
ist alone, especially in first class theaters. However, even where a 
service engineer makes regular calls to the theater and is readily avail- 
able in any emergency, the projectionist still can and should do a 
large and important part of service work; not the finer points re- 
quiring elaborate test equipment and special training, but the funda- 
mental work which is still much the same as it has always been. Then, 
too, he can and should be more familiar with the arrangement, wiring, 
and individual peculiarities of his own installation than it is possible 
for anyone else to be. 

Properly kept maintenance records of the sort previously discussed 
are the backbone of good servicing. They provide most of the in- 
formation necessary for discovering and correcting weaknesses before 
they actually develop into trouble. The remaining part of servicing 
concerns locating and correcting trouble that happens unexpectedly. 
In other words, "trouble-shooting." 

We usually think of this job as beginning in a wild scramble when 
the show stops or has to be stopped during a performance. Indeed, 
that has been the usual procedure in most projection rooms, and it is 
nothing to be proud of. If the show has been restored quickly by 
such procedure it has been purely a matter of luck. 

At least nine-tenths of good trouble-shooting should be done be- 
fore trouble becomes apparent in results. The first requirement is 
to know exactly where each electrical circuit originates and ends; 
through which fuse blocks, switches, terminal strips, junction boxes, 
and equipment parts it passes; what voltage and amperage it should 
carry; whether a-c or d-c; and if d-c, the polarity of each wire. 
As much as possible of this information should be plainly written on 
a label or tag which is then permanently attached to the equipment 
or to the wires themselves at every point where they can be reached 

148 J. R. PRATER [J. S. M. P. E. 

conveniently for testing. In addition, drawings of the internal 
wiring of each piece of equipment should be obtained from its manu- 
facturer, unless already on hand. Always check, or have the engineer 
check such drawings with the actual equipment and wiring, and 
immediately make any changes necessary. Then either attach the 
drawings permanently to some convenient part of the equipment they 
illustrate, or record them in the maintenance record book. 

The next step is one that should be obvious to anyone who is even 
remotely connected with a theater. We all know that for any per- 
formance to go on smoothly it must be rehearsed, or at least carefully 
planned in advance. If the performance of trouble-shooting is to go 
forward swiftly and efficiently, it must also be rehearsed and planned 
before actual trouble is expected. 

Projection equipment, and especially its installation, varies too 
widely for any general outline to be entirely satisfactory. All the 
information necessary to locating and correcting (assuming that it 
be possible) any trouble the projectionist is likely to encounter is 
easily available to him in manufacturers' instruction books, good 
textbooks and magazines, or from the service engineer. Even de- 
tailed step-by-step charts of exactly what to do in the most logical 
order of procedure for almost every known type of trouble in the en- 
tire theater sound system have been published, both in magazine 
and in textbook form. 

The duty of the projectionist, or at least of the chief projectionist 
of each theater, is to use this material as a reliable guide, and to con- 
struct a similar written outline covering everything he can do in his 
own individual case, but nothing more. The next job, and a most 
important one, is to go over his entire installation of equipment very 
thoroughly step by step, actually making every possible test that he 
could make if trouble existed. In fact, whenever possible, without 
causing damage, he should deliberately create the trouble and then correct 
it. Since many of the possible troubles may never occur, or if so only 
at infrequent intervals, it is impossible to remember all the details 
of each test. Therefore it is necessary to combine with the outline a 
written record of the exact results obtained in each case with the 
actual test equipment available to the projectionist. This is very 
important, because his methods and test equipment may give en- 
tirely different results from those of an engineer, but as long as he can 
repeat the exact test in the future and compare the results with those 
of the same test made when the equipment in question was in normal 


working order, he can usually tell good from bad and that is usually 
all he needs to know. For example, he can determine accurately at 
any time whether or not the contact-pressure of each brush in a 
motor-generator set is correct, with only a rubber band and a weight 
of any kind equivalent to the proper contact-pressure, provided he 
has prepared the weight when he knew the brush-pressure to be cor- 
rect. Similarly, he can show his manager how badly the screen has 
deteriorated by comparing visually its present reflection power with a 
sample of the same material that has been carefully wrapped and 
stored since the screen was new probably more convincingly than 
a competent engineer with a similar comparison of light-meter read- 
ings could do. The projectionist can take all his amplifier tubes, 
including spares, to the local supply dealer for testing, and can record 
the meter reading obtained for each tube. Even though the readings 
may be calibrated in numbers that mean nothing except varying de- 
grees of excellence or the lack thereof, they provide a reasonably satis- 
factory means of choosing between good and bad tubes, as well as of 
matching them in pairs for use in half-wave rectifiers or push-pull 
amplifier stages. 

The projectionist who keeps accurate written records as above sug- 
gested, and who makes all possible tests and emergency hook-ups, 
both mechanical and electrical, while he knows his equipment is in 
good working order, will seldom encounter a breakdown that he can 
neither correct nor side-step with a temporary repair until replace- 
ments or service can be obtained. Even when he has exhausted his 
own resources without success and must send for an engineer, he can 
make preparations that may save valuable minutes when help arrives, 
especially if there is any chance of making repairs soon enough to save 
dismissing the audience. For instance, if the projectionist has defi- 
nitely located a faulty unit, he can remove it and be ready to install 
the new one as soon as it arrives. If further tests must be made, he 
can remove any panel covers or parts that will prevent or obstruct 
free access to points that must be reached, get out all available draw- 
ings, blueprints, and reference data associated with the trouble and 
look up any specific information he can that the engineer may need 
upon his arrival. If a soldering iron will be needed the projectionist 
should have it hot and ready a few minutes before the engineer is 

It is granted, without argument, that professional service by a com- 
petent engineer with elaborate test-equipment is a necessity in ob- 

150 J. R. PRATER Q. s. M. P. E. 

taining the last bit of perfection in screen results from modern pro- 
jection equipment, as well as in locating and correcting the more 
difficult cases of trouble. But in the thousands of theaters where 
such service is a luxury that can be enjoyed only at rare intervals, as 
well as in the thousands more where it takes hours or even days to ob- 
tain replacements or outside service, the patrons who pay their ad- 
mission fees are entitled to at least a reasonably excellent performance. 
The responsibility of delivering that performance rests largely and 
primarily in the hands of the projectionist. The results he produces 
depend less upon how elaborate his resources are than upon how well 
he uses those he has. 


Laboratory technicians of widely varying kinds have combined 
their efforts to give us the motion picture as it is today. The results 
of their efforts may be seen in almost every branch of the entire 
motion picture industry. 

Photography has been developed tremendously. The very film 
upon which the pictures are made has been changed from a brittle, 
yellowish substance to one that is tough, pliable, and almost wholly 
colorless. Early photographic emulsions capable of producing pic- 
tures only slightly better than silhouettes, and even then only under 
the most favorable conditions, have been replaced with dozens of 
special emulsions, each capable of doing a certain difficult job. 
Numerous color processes have been combined with, or added to, 
these new films and emulsions. Camera mechanisms and lenses have 
been improved and redesigned to take full advantage of the almost 
endless possibilities furnished by modern raw film stocks. Develop- 
ing solutions and processing methods have kept in step. Without 
the single process of "duping" alone, modern quantity production 
would be impossible. 

Sound recording for motion pictures has undergone similar exten- 
sive development and improvement. Disk records with their needle 
scratch and synchronization problems have been replaced by direct 
recording on the film itself. Slowly but steadily the fidelity and 
frequency response of sound recording has been improved, while 
undesirable noises of all kinds have been almost eliminated. Re- 
recording has solved difficult editing problems. Uniform standards 
have been agreed upon. 

Projection once consisted solely in getting an image of any kind on 


the screen. Lenses were slow and poorly corrected, intermittent 
movements were inaccurate, mechanisms as a whole afforded little 
or no protection to the film, often inflicting serious damage them- 
selves. Projection speed varied from half to twice normal camera 
speed according to the projectionists' individual fancy or to a more or 
less unreasonable performance schedule. Light-sources were weak, 
and their optical systems could deliver to the aperture only a small 
portion of what light was produced. Modern projectors and their 
high-intensity arc lamps are capable of delivering to the screen an 
almost rock-steady, sharply defined, undistorted, and brilliantly 
illuminated image at a projection speed exactly equal to normal 
camera speed. More than that, the job can now be done efficiently, 
dependably, safely, and with a very minimum of wear on the film. 

Sound reproduction has followed closely behind the advances 
made in recording. It has grown from the rank of a mere novelty 
to that of an absolute necessity to theatrical motion pictures. Equip- 
ment and methods have been improved steadily, and at the same 
time simplified, until even the smallest theater can now reproduce 
any sound from a whisper to a concert by a full symphony orchestra 
with remarkable naturalness. 

All these developments and refinement that make the motion pic- 
ture of today a technical marvel of almost perfection in every detail 
have come from laboratories. But research, experimentation, 
special equipment, and the services of thousands of highly trained 
technicians have cost a tremendous amount of money. Without 
that money, practically none of this development would have been 
possible. The motion picture would still be a crude novelty. 

Whence has all this money come? It has come from the patrons 
who have paid their admission fees to motion picture theaters. Not 
only the money for improvement, but also that for production, dis- 
tribution, and exhibition, as well as what is wasted or lost along the 
way, has come from the theater box-office ever since the day of the 
first "nickelodeon." The total revenue from all other sources com- 
bined is but a drop in the bucket. 

Between what the laboratory has made possible and what the pay- 
ing customer actually sees and hears in the theater, there is a gap 
filled by the projectionist. The final success or failure of the efforts 
of the entire industry depends largely upon how well or how poorly 
the projectionist does his work. It is true that if he puts on the show 
at all, some of the benefits of the careful work already done by tech- 

152 J. R. PRATER [J. s. M. p. E. 

nicians will reach the patrons. However, it is also true that the 
finer those benefits may be the more easily they can be lost before 
reaching their final goal. We know, of course, that no projectionist 
can produce better results than the films and equipment furnished 
him are capable of, but let us suppose a new print is to be projected 
with modern equipment in first-class mechanical condition. Here 
are a few items that would depend even then upon the projec- 

An error of a few thousandths of an inch in focusing the projection 
lens, or a fingerprint or smear of oil on any of its glass surfaces will 
reduce the finest photography to something definitely displeasing. 
Too little aperture tension or a tiny bit of dirt or emulsion lodged 
on the face of the intermittent sprocket will cause the screen image to 
be unsteady in spite of the fact that the tolerances of the intermittent 
movement have been held to within one ten-thousandth of an inch. 
Too much aperture tension will cause unnecessary wear and inflict 
permanent damage to the film. A slight error in any of the arc- 
lamp adjustments may reduce screen illumination, create uneven dis- 
tribution of light, cause unwanted color to appear, allow the light to 
fluctuate, or even to go out entirely. A tiny bit of lint or dirt ob- 
structing part of the light-beam of the sound optical system may 
impair sound quality more than the last ten years of work have been 
able to improve it. The same result may come from an improperly 
adjusted exciter lamp, or from dirt or oil on sound optical system 
elements. A small accumulation of wax or dirt on the face of the 
sound-sprocket or film-drum may introduce enough flutter to make 
the sound displeasing to even an average listener. Too little volume 
may rob the sound of intended dramatic effect and naturalness, or 
make parts of it wholly inaudible. Too much volume may also 
make certain parts of sound unnatural, and the louder parts distinctly 
annoying or even painful to many ears. A few runs through oily, 
dirty projectors and an improperly adjusted rewinder will inflict 
permanent damage to the film, which will result in blemishes in the 
screen image and noise in the sound not only in that theater, but in 
every theater where that print must be run. 

These are but a few of the possible instances in which the results 
of years of the most careful laboratory work may be destroyed or 
may fail to reach the theater patrons for whom they were intended, 
through failure of the projectionist to do his full share of the job. 
Such failure may be due to many causes, but outstanding among 



them is the lack of thorough technical knowledge. This is the very 
backbone of ability and pride in good workmanship. 

Whatever can be done to increase the technical knowledge of the 
projectionist will benefit the entire motion picture industry just as 
much as it will the projectionist himself. It is, of course, impossible 
for the projectionist to know as much about every phase of the work 
as do the engineers, who usually specialize in some one of the many 
branches that compose the entire field The best that may be hoped 
is that he may be able to understand the fundamentals of all branches, 
and enough of the newer developments in each to be able to under- 
stand and appreciate what the engineers are doing and to cooperate 
with them to the extent of handling their productions competently. 

If the projectionist is to do this and thus deliver laboratory results 
to his audiences, he must be able to understand if not to speak the 
language of the laboratory engineer. A spirit of friendly cooperation 
must exist between them if the final result of their well directed 
efforts is to attain final maximum success. 



Summary. The high importance of expert work in theater projection rooms is 
stressed. Pride in performance is essential to excellence, and if the status of pro- 
jection were elevated to a higher plane improvements would result in both picture and 
sound presentation. The author offers a suggestion concerning the contacts of the 
Society with the projectionists' organizations. 

This paper has been prepared in the hope that its contents may 
start discussion eventually resulting in greater justice to the splendid 
equipment made available to theater projection rooms through the 
efforts of able engineers connected with their planning and construc- 
tion. It has been prepared in the hope that such discussions may 
result in better presentation of the product Hollywood delivers to 
theaters in the form of still photographs successively arranged, ac- 
companied by sound in photographic form, all for the entertainment 
of countless millions of men, women, and children each day. 

While this industry and its various component parts are slowly com- 
ing to realize that, if these successive still photographs are to be trans- 
formed into an acceptable semblance of the original motions and 
sounds they represent, great care, coupled with skillful manipulation 
of the equipment employed, must be applied in theater projection 
rooms. There still is lack of comprehension of how far we are from 
securing maximum excellence on theater screens and through loud 

The design engineers have worked wonders in planning and con- 
structing these equipments; yet there is a sad lack of care in their 
use in the theaters. This is not due entirely to lack of knowledge, as 
ample information is available to projectionists. Nor is it due to lack 
of skill, but rather to lack of careful attention and application. 

In the theater of today, great and small, well equipped or other- 
wise, we find a large amount of what can only be termed carelessness. 

* Presented at the 1939 Fall Meeting at New York, N. Y. ; received October 
10, 1939. 

** Quigley Publishing Company, New York, N. Y. 


We find projectionists little interested in the finer points of their work. 
We find few cases where the screen image is hi the sharpest possible 
focus. We find the screen poorly illuminated, with not a word of 
protest from the projection staff. We find great waste of light which 
might be avoided by care and skillful work on the part of the pro- 
jectionist. We find sound that might be greatly improved; and so 
on through a considerable list of remediable faults. 

A course of education in technic would remedy much of this, and 
would be a most effective procedure by the Society. However, that 
by itself would not be sufficient. 

I have visited many theaters hi many parts of the country, and 
have had a unique opportunity to ascertain what is wrong in theater 
projection rooms. The chief fault is, I should say, a lack of pride on 
the part of projectionists hi the performance of their work as exempli- 
fied by the screen and sound. It is a rare thing to come across a 
projectionist who honestly takes pride in the results of his work. 

Perhaps this is the result of the indifference shown toward and the 
lack of respect exhibited for projection and projectionists through 
all but a very few of the recent years of our industry's existence. I 
firmly believe that great improvement would result were the pro- 
jectionist and his work raised to a higher plane of respect. What 
can our Society do to assist in such a movement? It might, at 
its meetings and in its publications, direct attention to the great need 
of careful, expert cooperation in putting on good motion picture 
shows. This might tend to raise the respect of the projectionist for 
his work and arouse his pride in the work. It might be well to con- 
tact the IATSE through its international officers and perhaps its 
conventions, stressing the point that it is the duty of the local unions 
not only to supply to exhibitors men who are thoroughly competent, 
but also to check up on their performance and see to it that the best 
possible results are produced by its members. 

It would seem, too, that the exhibitors' organizations might well 
be contacted by our Society, calling attention to the improvement in 
results in both picture and sound, and the reduction in cost resulting 
from employment of men of higher ability possessed of greater re- 
spect for and pride hi performance. Set up a condition where men 
and their work are held in high respect, and automatically those men 
will react favorably. Set up a condition where men and the results 
of their labor are held in small respect, and there is little chance that 
excellence of performance will result. 



Summary. Sixteen-mm sound motion pictures are potentially one of the most 
effective means through which industry can develop a broad, cost-cutting communica- 
tion system within the organization itself. 

Many latent applications for internal films exist; the cases in business where the 
improved transfer of ideas afforded by films can be most profitable are almost un- 
limited. Several specific instances are cited. 

Sixteen-mm equipment is simple, easy to operate, reliable, and economical. 
With it, a member of the industrial organization who knows his company's products, 
policies, and structure can readily produce films that are, in every respect, profitable 
internal communications media. 

It is usually a surprise to even those connected with our industry 
to learn that the virtues of Dewar's Scotch Whiskey were extolled 
in a business film in 1894. 1 As we look further into the history of 
films for business purposes, we learn that business films are practically 
as old as the motion picture itself. 

Unfortunately the growth of the business film has not paralleled 
the growth of the industry. As one writer put it: "The point is, 
commercial motion pictures are at least 45 years old ; yet advertisers 
still think of them as a new medium. Here is a major medium of 
expression that has grown, to be sure, but has never grown up. 
Why?" 1 Many answers to this question have been suggested, a 
number of which can be traced to the same underlying cause : "Films 
are regarded at the foot of novelty advertising, below souvenir 
pencils and paperweights instead of at the top of communications." 1 

It should be obvious that any communication medium is particu- 
larly fitted to specific classes of intelligence transmission; this prin- 
ciple applies no less to the film than to any other communication 
medium. As we review the films produced by education and those 
produced by business, we are struck with the similarity of the amis of 

* Presented at the 1939 Fall Meeting at New York, N. Y. ; received October 
14, 1939. 

** The Berndt-Maurer Corp., New York, N. Y. 


the two fields and the dissimilarity of the films produced. We are 
tempted to conclude, if a cryptic summary may be ventured, that 
business has produced much film and little theory as to how films 
should be made and used, whereas education has produced much 
theory and few films in accordance with the established theory. 
Since the aim of both films is to "put an idea across," it is reasonable 
for us to extrapolate the experience of one into the field of the other 
to the mutual benefit of both. 

One of the simplest and best reports 2 on the subject of films for 
educational purposes is that of the Committee on Intellectual Co- 
operation of the League of Nations, written in 1924. The findings 
of this Qommittee may be considered a reference standard through 
which the efficiency of instructional films may be gaged. The 
following is a resume" of the report. 

(1) Slides and motion pictures should be used for maximum 
effectiveness. As a general rule the use of these two adjuncts is not 
judiciously proportioned, one oftentimes being used to the complete 
exclusion of the other. We can no doubt agree that the conclusion 
of fifteen years ago is still a valid one today. Too many projects 
use but one medium to the complete exclusion of the other. 

(2) All objects and scenes that the audience is intended to watch 
and remember in movement should be shown in movement. Still 
pictures representing objects and scenes that ought to be seen in 
movement should be banned as giving a distorted impression of 
the actual facts. While our films are improving in this regard, we 
still find countless instances where we photograph stationary objects 
with a motion picture camera and moving objects with a still camera. 

A corollary that may reasonably be added at this point is that all 
objects and scenes that the audience is intended to watch and re- 
member hi sound should be shown in sound. Silent pictures repre- 
senting objects and scenes that ought to be heard in sound should 
be banned as giving a distorted impression of the actual facts. 

(5) The screen can not displace the personal element; it can to 
some extent displace printed matter, and it should, in all events, be 
used in combination with it. The use of the screen in conjunction 
with text-books and printed matter is still quite undeveloped; its 
effectiveness when properly used is in the top rank of communication 

(4) The screen should be used in combination with personal con- 


tact in "getting the idea across." It should be used at the location 
where the salesman or teacher ordinarily operates whenever it is of 
advantage to do so. It should be possible to repeat the picture several 
times if necessary; the picture should be definitely constructed in 
such a manner that it will bear repetition. 

Industry has shown a growing tendency to bring the screen to the 
customer instead of the customer to the screen. This tendency is in 
the proper direction. 

It is not true, however, that films are always constructed in such 
a manner as to bear repetition, as is particularly necessary in films 
for instructional purposes. Too often a large number of diverting 
technical effects such as fancy wipes, dissolves, and the like, have been 
used in a single reel. Such technical effects do not cover up glaring 
defects in plot, continuity, and lack of logical presentation that are 
also usually present. Our technical effects shall aid the story, not 
make it. 

(5) The screen can not be used in the proper manner unless there 
is very wide distribution of effective yet inexpensive apparatus, so 
that every user of films can have his own projection equipment. The 
simplest apparatus to handle will be best, and at the same time there 
must be no risk of fire. If the screen is to do its proper work the 
apparatus must quickly become a thing in daily use. 

The 16-mm size affords the best means of very wide distribution 
of effective yet inexpensive apparatus with minimum risk of fire. 
It is the simplest to handle and makes possible the daily use of the 
equipment due to its very simplicity and low cost of operation and 
maintenance. That this is true of equipment for making films as 
well as of equipment for reproducing films is only now beginning to 
be appreciated in its broader aspects. 

(6) The mode of use of the screen must be improved, having 
regard to the fact that it can act upon the mind of the spectator: 

(a) By faithful presentation of the subject. 

(b) By the representation of the subject simplified. 

(c) By the representation of the subject in sections. 

(d) By the representation of the subject intensified, magnified, represented, 
speeded up, slowed down, built up by degrees, or superposed. These different 
methods must be employed according to a logical scheme, taking into account the 
subject to be dealt with and the specific character of the audience to which the 
film is planned to be shown. 

Feb., 1940] 16-MM EQUIPMENT IN INDUSTRY 159 

"Too often they (the producers) have sold companies on the idea of 
producing films, not as integrated parts of a well rounded-out pro- 
gram, but as special bits of magic for one-time splurges." 1 Even in 
the larger efforts it is almost as common for a film to be designed for 
no particular audience as it is in the case of the film produced by the 
fly-by-night "Hollywood" director who "has his office in his hat." 
There is still the feeling among a number of picture purchasers, and 
to a lesser degree among picture producers, that one magic super- 
spectacle is better than a large number of modest films each telling 
its complete part of an integrated story. 

It is of utmost importance that the exhibition plans for a film be 
fully completed before the first camera exposes the first foot of film. 
Maximum effectiveness presumes the gearing of the subject matter 
of the film to the audience. 

(7) The screen is a valuable means of suggestion; it will be used 
as a time-saver, often a valuable one, in "putting across" all matters 
that depend largely on visual memory. 

Psychologically, the lighted screen in the darkened room compels 
concentration upon the material presented. It is not only possible 
to "put across" details of mechanisms and their operation, but also 
to explain the coordination of the activities of groups that can not in 
the usual course of events be observed. This field is practically a 
virgin field for industry. 

(8} In order to economize effort and to save expense in making 
films, and to derive maximum profit from them, it is advisable to 
decide definitely beforehand to what extent regular photographing 
and animation are respectively to be used. 

Due to the high cost of animation per foot in comparison with 
regular photographing, animation is used to a much smaller degree 
than in many cases seems desirable for maximum effectiveness. 

If we reexamine the field of business films as a whole, we are struck 
with the fact that the external film, in particular, the film developing 
the customer-sales-organization relationship, has been widely used. 
The internal business film is less widely used and still less widely 
heralded. Many organizations that have used both have found that, 
dollar-for-dollar, the second type usually produces better results 
wherever it has been tried. The obvious use of the internal film is 
in sales training, and many organizations have set up photographic 
departments to make use of the advantages of this type of film. 


The well-managed industrial organization is always looking for new 
opportunities to achieve and maintain competitive advantage. 
When a unit for the production of internal business films is first 
organized, its personnel complement is small and its activities few. 
These few activities are usually related in some manner or other to 
the selling operation. But when the film production unit gets under 
way, its scope of activity expands far beyond its original purposes 
into fields that have little to do with the sales operation per se. 

An excellent example of this is a recent film produced by the Fisher 
Body Division of the General Motors Corporation showing how 
container-packed automobile body parts may be advantageously 
handled by transportation companies between the manufacturing 
plant and the assembly plant, to the simultaneous profit of both the 
transportation companies and the Fisher Body Division. This film 
was produced wholly within the company organization. While it 
was necessarily photographed in "catch-as-catch-can" manner, the 
resultant film tells its story forcefully as well as effectively and needs 
no technical embellishment whatever to establish the straight- 
forward points of the presentation. 

The establishment of such internal motion picture departments in 
industry is now becoming quite common. Some organizations, such 
as the Fisher Body Division, in the case just cited, prefer to produce 
the film completely within the organization. Other organizations, 
such as the Skelly Oil Company with its salesman-training films, 
prefer to work out the script and shoot the basic material, leaving 
the editing and scoring work in the hands of a commercial 16-mm 
film producer. The proper procedure depends upon the circum- 
stances; each has its respective advantages. The 16-mm industry is 
prepared to supply not only all the necessary equipment but also all 
production and other services required for all such needs. 

The advantages of 16-mm equipment for internal production pur- 
poses hardly need repetition here. In 16-mm production equipment 
as in projection equipment, relative simplicity, portability, freedom 
from fire risk, and relatively long operating time per pound of film 
are already well known. The picture quality and the sound quality 
may be readily made as good as required. In the case of the sound, 
for example the quality may readily approach that of 35-mm 
theater reproduction if desired, as was demonstrated at the last 
meeting of the Society by J. A. Maurer. 1 In most cases such high 
quality is not required. 

Feb., 1940] 16-MM EQUIPMENT IN INDUSTRY 161 

The subject matter of internal films does not require treatment by 
experienced showmen and dramatists; it is best presented in straight- 
forward expository style. Subject matter need not be created and 
staged; there is an abundance of it to be found in the plants, in the 
men, and in the routines of the organization. 

The ultimate in technical excellence is not required to enable these 
films to serve their purpose effectively. If the picture is reasonably 
well exposed and correctly focused it makes little difference that the 
scene was not lighted by artists. If the sound is natural and under- 
standable, and properly related to the subject, it makes little dif- 
ference that it is not the masterpiece of a showmanlike mixer-man. 
In such films the subject matter is of predominant importance. 

The all-important qualification of a producer of internal films is 
an intimate knowledge of the organization for which he intends to 
produce. He must be thoroughly familiar with its policies, its plans, 
and its objectives. He must be on a firm footing with the personnel ; 
he must know executives and managers and have their full confidence. 
Most of all, he must know and understand the various problems 
existing in the company. Only with qualifications such as these can 
he expect to make his work appear convincing and authentic. 

It would seem difficult to find men answering these requirements 
anywhere except among the company's seasoned employees. Can 
such employees become the producers of internal organization films? 
The answer is definitely "yes"; the required technical ability is 
readily developed, as anyone familiar with the characteristics of 
direct 16-mm camera and sound recording equipment can understand. 
The technic of the medium is, as a general rule, readily acquired by 
a person who has the other necessary qualifications. 

There are two general classifications of business films for internal 
use: those that convey a message from management to personnel, 
and those that convey a message from personnel to management. 
The films in the former group are primarily instructive; they may 
therefore be expected to follow the usual four-step instructional 
method of preparation, presentation, application, and examination. 
Other media of instruction, such as the lecture, the slide-film, and 
the printed word, should be integrated parts of the instruction 
method. Since the films in the latter group are primarily informative, 
the examination step usually is not involved in the presentation. 
The use of other media of communication such as the lecture, printed 

162 W. H. OFFENHAUSER, JR., AND F. H. HARGROVE [J. s. M. p. E. 

Feb., 1940] 16-MM EQUIPMENT IN INDUSTRY 163 

word, slide-film, and so forth, is also indicated here just as in the 
case of instructional film. 

Fig. 1 shows a general outline of films for internal business use. 
Certain of the applications shown have been widely used; others 
are still relatively little developed. Most of the classifications can 
be readily understood by inspection and need little explanation. 

A few notes on some of the lesser understood classifications are 
indicated at this point. These notes deal with the management-to- 
personnel films of the non-sales-service group. 

Interdepartmental Organization. In larger organizations particu- 
larly, the loss of the personal touch is a morale factor that should 
engage the attention of every business manager. Specialization 
makes such demands upon the time of a particular individual that 
it is not only impracticable but usually also impossible to maintain 
that desirable form of contact. There are numerous companies today 
where a film outlining the organization of the company would not 
only reestablish the personal contact but also depict clearly the per- 
sonalities of the men performing the various functions in the organiza- 
tion. This type of film can materially aid the esprit-de-corps and im- 
prove the efficiency of the organization. 

Safety Promotion and Health Conservation. The loss of time and 
efficiency of personnel due to sickness and accidents is still a major 
problem to industry. Most industrial plants have safety and health 
programs, which vary in scope from weekly bulletin board cartoons to 
elaborate systems including medical service, health furloughs, and 
other features. In industrial plants particularly, it has been found 
that motion pictures are the most graphic means of safety and health 
education. The work that is now being done in connection with 
safety training in the mining industry is a typical example. The 
motion picture makes it possible to show clearly all the details of 
the little items of carelessness that result in an accident. 

Job Technic Training. Job technic training is one of the foremost 
non-sales uses for internal films. It is not unusual to find that 
older employees are poor teachers, and, because of possible jealousy 
of new employees, such older employees may even intentionally 
pass on instruction that is not of the best. The film admits of stand- 
ardized instruction into which error through repetition can not creep. 
This standardized instruction may readily be that of the most effi- 
cient method of performing a particular task. 

New technics must be acquired by old employees, not only to im- 


prove their productivity on old products but also to produce new 
products. If an employee is left to devise his own technic, time- 
and-motion study also best done with motion pictures will show 
that he usually develops a definitely inferior technic. 

Employee Relations. Employee relations is a wide subject that 
takes in practically all matters that improve the feeling of the em- 
ployee toward his company. It is now recognized as good manage- 
ment practice to do whatever is possible toward making each em- 
ployee feel that he is important to his company, and that his work 
makes him an important member of society as well. 

The ways in which films can contribute to the improvement of such 
employee relations are almost limitless. Films can be made that give 
direct educational treatment to such subjects as the aims and prob- 
lems of the company as well as the soundness of our traditional 
economic system. Films can be made to provide entertainment for 
specific employee groups, catering to their particular tastes. Films 
can be made of employee outings, athletic events, and social events. 
Films can be made of a documentary or newsreel type depicting 
existing technics and dealing with local conditions. Such films may 
be used for historical purposes or for comparing the employee ac- 
tivities in one plant with those in another. Most executives engaged 
in personnel work can readily visualize a host of other applications. 

Employee Advancement. In progressive organizations it is a 
cardinal principle of management that anything that increases the 
intrinsic value of an employee automatically increases his value to 
the company. In line with this management principle is the or- 
ganization principle that a man who fills a particular position should 
gradually acquire wider and wider experience in the work of the 
position directly above him. In this manner each man in the or- 
ganization is progressively relieved of more and more routine work 
by capable, trained assistants. This condition makes for organiza- 
tion flexibility in that it not only makes possibje the replacement of 
personnel losses resulting from ordinary causes such as sickness and 
death and the usual labor turnover, but also the replenishment and 
even expansion of personnel required by market and product expan- 
sion, or by emergency causes of whatever nature. 

In employee training for advancement, as well as in job technic 
instruction, the 16-mm sound motion picture finds a logical applica- 
tion. The initial source of basic film material may be film produced 
for other purposes but edited into an appropriate version, or it may 

Feb., 1940] 16-MM EQUIPMENT IN INDUSTRY 165 

preferably be a film made especially for the purpose. Inasmuch as 
employee advancement must ordinarily be carried on outside of 
regular business hours and in addition to the usual duties and routine 
of the employee concerned, it is doubly important that a wide range 
of essentials be covered in the most effective and the most efficient 
manner. Sound-films offer a means of greatly reducing the time 
allotments for the presentation of facts without sacrificing the quality 
of instruction. 

In almost every broad generalization concerning films, some notable 
exceptions to the general rule can usually be found. In the field of 
training films, one such exception is the work of the Photographic 
Section of the U. S. Army Signal Corps. In attacking this film 
problem with its characteristic thoroughness, the Army found that 
no deviations from the fundamental rules governing training-film 
production can be tolerated. Each film must be specifically pre- 
pared for a particular audience and every effort made to avoid 
entertainment features or to produce a film suited to what is often 
called "the general audience." A number of papers *- 5 6 on the subject 
of the Army training film program have been presented to our Society 
and all are worthy of very thorough study by anyone concerned with 
personnel training. The Army demonstration films shown before our 
Society prove that the practical development of the training-film has 
reached an advanced stage quite beyond anything done elsewhere on 
a scale of appreciable scope. 

The second notable exception is best described in an announcement 
(Feb. 10, 1939) from the Harvard Film Service of the Biological 
Laboratories of Harvard University, concerning film material for the 
improvement of reading: 

"The Harvard Film Service in cooperation with the Psycho- 
Educational Clinic, Harvard University, announces a new type of 
film material for the improvement of reading. 

"In brief, these films consist of reading material so presented that 
successive phrases of the separate lines are exposed rapidly across 
and down the screen. The film serves as a 'pacer' and the pupil is 
stimulated to keep up with the rate of exposure. As the training 
progresses, selections with longer and longer lines are presented 
thereby gradually increasing the eye span. 

"During the first half of the current academic year, these films were 
tested out in an experiment on a group of slow readers among Harvard 

166 W. H. OFFENHAUSER, JR., AND F. H. HARGROVE [j. s. M. p. E. 

freshmen. The group met for a 45-minute training period three 
times a week for eight weeks. The results were as follows: at the 
close of the experiment, the trained group averaged gains of 41 
percentiles on a speed-of-reading test( the Minnesota Speed-of- 
Reading Test for College Students) and of 24 percentiles on a test of 
accuracy of reading (Whipple's High-School and College Reading 
Test) in excess of those made by a non-trained control group. When 
measured in terms of a difference between initial and final eye-move- 
ment records, the average gain in rate of reading of the former was 
52 per cent. An analysis of these records hi terms of individual 
measures showed that the average number of fixations per line was 
reduced from 10.8 to 6.5; the average number of regressions from 
1.6 to 0.5. 

"This material is designed to be run at silent speed (16 frames 
per second) on any 16-mm projector. It may be used for a single 
pupil or for a group and requires only a semi-darkened room. Twenty 
selections averaging 125 feet each, adapted to the senior-high and 
college levels, together with a teacher's manual and a set of com- 
prehension tests for each film, will be ready for release on March 
first. . . . Although a smaller number of films may be purchased, the 
best results will be secured when the complete series for any given 
level are used. By April first, in time for a two months' training 
period this year, we shall have ready for release thirty selections for 
Grades 3 to 5; by next September, a third set for Grades 6 to 9." 

Life is daily becoming more complex not only in its social aspects 
but also in its business aspects. It is imperative that our training 
methods become more effective from the standpoint of saving time 
and improving quality of instruction. In addition, training of ever- 
increasing scope must be made available to an ever-expanding group 
of our trained citizens. Business has already begun to show an eager 
interest in the development, and if present indications can be relied 
upon, it should not be very long before we shall be hearing of the 
experimental results of applied programs of internal training de- 
velopment such as we now suggest. 


1 "Camera Action Sales," Report to Executives, Business Week (May 27. 
1939), p. 38. 

* SEABURY, WM. M. : "Motion Picture Problems The Cinema and the League 
of Nations," p. 248. 

Feb., 1940] 16-MM EQUIPMENT IN INDUSTRY 167 

3 MAURER, J. A.: "The Present Technical Status of 16-Mm Sound-Film," 
/. Soc. Mot. Pict. Eng., XXXIII (Sept., 1939), p. 315. 

4 HOORN, F. W.: "Military Training and Historical Films," J. Soc. Mot. Pict. 
Eng., XXI (Oct., 1933), p. 337. 

6 GILLETTE, M. E.: "The Use of Films in the U. S. Army," J. Soc. Mot. Pict. 
Eng., XXVI (Feb., 1936), p. 173. 

PROSSER, W. E.: "Motion Picture Activities of the United States Army," 
Trans. Soc. Mot. Pict. Eng., No. 38 (May, 1929) p. 355. 


Commission on Educational and Cultural Films, "The Film in National Life," 
George Allen and Unwin Ltd. (London), 1932. 

DEVINE, J. E.: "Films as an Aid in Training Public Employees," Report to 
the Committee on Public Administration, Social Science Research Council. 


MR. BRADY: Has industry taken to the motion picture to a large extent? 

MR. HARGROVE: There are today forty-eight fairly large industrial companies 
who are making their own motion pictures, most of them in the 16-mm size. 

MR. BRADY: What prevents industrial companies from seeing immediately 
the advantages of the 16-mm film? 

MR. HARGROVE : Probably the lack of understanding of what can be done with 
the medium, and the general idea that a motion picture has to be entertaining, 
rather than educational or instructional. 

MR. BRADY: Is the cost or the lack of projection equipment an important 

MR. OFFENHAUSER : Not as much as a better understanding of the principles 
involved. The films mentioned in the Harvard announcement, for example, 
are films that could be made with the cheapest of 16-mm cameras. 

MR. MITCHELL: Quite a number of industrial concerns have started to use 
16-mm motion pictures because someone in the company had produced or de- 
signed something new, and wanted a record of what he had done. The first thing 
he knew, the sales department borrowed the film, and it began to attract the in- 
terest of others in the company. Many executives of the big companies do not 
know or appreciate how this new tool can be used. 

MR. GILLETTE : One of the principal hurdles that this type of film will have to 
get over is to develop a new technic for presentation, differing materially from the 
technic of entertainment films or the pure advertising film. The technic of making 
successful instructional films differs greatly from that of making the entertain- 
ment variety. 

If instructional films are to be effective they must go further into details, 
avoid distractions such as extraneous material, comedy, emotional appeals, and 
unrelated scenes, and conform to the basic principles covering the preparation 
of instructional material. Also the methods of using such films should conform 
to educational practice. 


Summary. A 200-foot reel and tray developing machine with multiple trays, 
motor driven reels, and manual reel transfer, is described. Fresh processing solu- 
tion is used with every reel, the amount depending upon the footage of the reel. 

Difficulties encountered in development and construction are described, and the ad- 
vantages evident in practice are given as one-man operation; solution economy; clean, 
energetic development, with linear exposure-density relation; uniformity of results 
with any quantity of film from 1 to 200 feet; no undue aerial or chemical fog; cleanli- 
ness; and flexibility in use or in extension to future developments. 

Intelligence of the highest degree, accumulated experience, and no 
inconsiderable amount of money have produced the remarkable 
equipment necessary for the accurate processing of motion picture 
film in great quantities, but little attention has been given to the 
needs of the many small laboratories scattered throughout the United 
States and other countries. The total footage processed by these 
small laboratories must be considerable, but their work rarely war- 
rants the cost of even the smallest of continuous machines, or of 
maintaining the required large quantities of solution. 

Mechanical deficiencies and the impossibility of obtaining uniform 
agitation make the pan and spiral reel type of equipment imprac- 
ticable. Variations of the spiral reel system have been devised, but 
they require large quantities of solution. 1 The rack and tank system 
still survives with all its inherent faults of air bells, rack marks, 
differential development with increasing depth of solution, impossi- 
bility of proper agitation, polluted and leaking tanks, etc. 2 ' 3 - 16 Most 
important, the amount of developer can not be reduced to just 
enough for a rack of film, so compensation for developer depletion 
must be made, a difficult problem even in a well equipped laboratory. 
Motor-driven racks of the looped film type have been devised, but 
again large quantities of solution are required. 4 

To overcome such handicaps, the following design requirements 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received Feb- 
ruary 6, 1939. 

** Municipal Light and Power System, Seattle, Wash. 


were indicated: speed, with ease and certainty of operation; no 
possibility of film distortion or damage; uniform and adequate agi- 
tation; the use of new solution for every charge of film processed, 
with solution quantity entirely according to the amount of film in 
each charge so that the depletion rate would be identical with any 
amount of film; and a capacity of 1 foot to 200 feet per charge. 
Certainly it is undesirable to cut a 1000-foot roll, but flexibility and 
size limitations make greater capacity impracticable. 

FIG. 1. The complete developing machine. 

A reel and tray machine meeting these design considerations has 
been built, and has given consistent and economical results. Fig. 1 
shows the complete machine, but not the standard type drying 
reel on which the film is rewound. There are five solution trays, all 
over a main tray draining to the sewer. Each tray has a hot and a 
controlled temperature water supply, and an independent drain to 
save solutions or to discard to the sewer. Each tray has bearing 
pedestals and a driving gear to engage the drum gear as it is dropped 
into place. 



[J. S. M. P. E. 

Fig. 2 shows the manual transfer of a drum from tray to tray by 
means of the lever mechanism carried along with the drum. Drums 
are moved from rear to front of the machine on a wheeled carriage. 
Fig. 3 shows the carriage in position for transfer of the drum to the 

A rectangular spiral wire soldered to the drum face guides the 
film so that winding is easily done in the dark. Very short lengths 
of film may be attached with adhesive tape, but it is necessary that 

FIG. 2. 

Showing the manual transfer of a drum from tray to tray by means 
of the lever mechanism on the drum. 

the film be kept tight on the drum, and the longer lengths are fas- 
tened with clips and rubber bands to take up the expansion. 

As expected in any development, difficulties were encountered in 
the construction of this machine. The reel and tray system 6 - 18 - 19 gave 
promise of solution economy and uniform agitation, but several in- 
herent faults had to be overcome. Trials with a wooden cage type 
of reel proved aerial fog to be at a minimum with fresh developer of 
either the borax or the positive type. A low-fog developer recom- 
mended for reel use, 3 ' 16 ' 18 and pinacryptol green 6 - 16 both increased 




200 FOOT 




5 ft! ?Ft. 

FIG. 3. Two-hundred ft reel and tray developing machine. 

172 R. S. LEONARD [J. S. M. P. E. 

rather than reduced fog. No attempt is made to explain this ab- 
sence of aerial fog unless the use of fresh solution is responsible. 

The paddle-wheel effect of the trial wooden cage drum caused 
uneven development. To eliminate this, and for solution economy, 
a closed type of drum was indicated. 

It was believed necessary that wash water reach the rear surface 
of the film, so a trial section of a fluted drum was built. This seemed 
satisfactory, but a full sized drum gave uneven development because 
of developer surging through the film perforations into and out of 
the fluting. 7 

Tests with a flat drum face showed that uniformity could be real- 
ized if the film remained flat on the surface so that developer could 
not circulate through the perforations. 

A smooth drum with a spiral film guide proved satisfactory, but 
expansion of the wet film, which was found to exceed 3 feet in a 200- 
foot length of acetate base, although much less with nitrate, caused 
overlapping of the center turns. Rubber bands took up the expan- 
sion at the ends but not at the center. A specially made soft-bristle 
brush held in contact with the revolving drum helped to carry the 
film expansion to one end of the drum where it could be taken up by 
the rubber band, but this could at best be only a temporary expedient. 

At a trial peripheral speed of 90 feet per minute, it was found that 
sufficient developer was not carried over with the drum and develop- 
ment streaks resulted. 

The drum speed was increased to 250 feet per minute, almost to 
the splash point. At this speed an appreciable quantity of developer 
carried around the drum, and uniform development was achieved. 
At this speed also, the film expansion carried to the drum end, and 
the overlapping of turns was eliminated. 

It was expected that this high film speed would cause directional 
degradation, 8 but examination, under a microscope, of heavily exposed 
areas adjacent to clear film showed no apparent trace of directional 
effect. Had this effect been evident, it was anticipated that periodic 
reversal of the drum would reduce or partially equalize the image 

Development tests with film strips exposed to a point light-source 
showed satisfactory uniformity of density, and no difference in de- 
velopment was evident at the center or at either end of the drum. 

It was expected that contact of the film base with the smooth 
drum would prevent removal by the wash water of traces of processing 


solutions. Examination of film dried on the drum showed no great 
amount of solids and it was planned to remove these traces in a sup- 
plementary bath combined with a squeegee. 

Clearance between drum and tray had been kept to a minimum, and 
it was found that one gallon of solution was just as effective in wetting 
the film as a great quantity. 

The high drum speed caused extremely energetic development, and 
it was found necessary to reduce the development rate. Dilution 
provided insufficient reduction, and dilution caused some developer 
constituents such as elon to be reduced to unreasonable amounts. 
Buffering the negative developer with boric acid in the proportion 
of 3 to 1 and reducing the alkalinity of the positive developer proved 
satisfactory. 9 ' 16 Sodium carbonate in reduced quantities does not 
follow a linear alkalinity rate, so Kodalk was substituted. 16 A satis- 
factory development rate was obtained, but it was found that a de- 
veloper compounded with Kodalk was sensitive to the aeration caused 
by the reel, and the development rate rapidly decreased indepen- 
dently of depletion. 

Tests were made of depletion rates and a decided difference in den- 
sity was found between the development of short lengths and the 
full 200-foot length. It had been anticipated that the spiral on the 
drum would cause adequate developer circulation, but tests with a 
dye solution placed in one end of the tray showed little circulation 
tendency. Two solutions to the problem were evident, developer 
circulation with a pump, or fractional footage reels. The latter was 
chosen and a 100-foot, a 25-foot, and a one-turn drum built. Other 
drums could have been made, but a reasonable balance had to be 
arrived at between cost and developer economy. The 100-foot size 
is useful for spring- wound camera rolls and for 16-mm if desired; the 
25-foot for shorter lengths and for printer-light tests; and the one- 
turn or 7-foot drum for sensitometric strips, exposure tests, or minia- 
ture camera rolls. Fig. 4 shows these drums mounted on a single 
shaft. A removable tray, divided into compartments to fit these 
individual drums, permits use of the proper amount of developer. 

Tests of depletion rates showed 100 feet per gallon to be a satisfac- 
tory compromise. This could be greatly exceeded in controlled 
practice with diluted solutions. The use of fresh solution seems to 
extend greatly the possible footage per gallon. 

The compromise between fractional size drums, and the rate of 
100 feet per gallon, has produced in practice no discernible differ- 

174 R. S. LEONARD [j. s. M. P. E. 

ence in density or contrast in development between 7 feet or inter- 
mediate lengths to 200 feet. 

The first two of the present five trays can be used for developer and 
short stop, or for two-solution development, 10 ' 16 the last two for in- 
creased washing capacity. The high drum speed with the small 
amount of wash water in the tray and its rapid rate of change, permit 
a high rate of hypo diffusion from the film with consequently shorter 
washing time. 11 ' 17 

FIG. 4. The 100-ft, 25-ft, and 7-ft drums mounted on a single shaft. 

Additional trays could be easily added if necessary for special 
work. The use of multiple trays makes possible uranium or other 
toning methods 12 and permits future processing of colorfilm if several 
solutions are necessary for the colorfilm of the future. Special treat- 
ment, such as intensification or reduction is also easily accom- 
plished. 13 - 20 

The use of smooth-surface white drums and a tubular or uniform 
light-source over the drum, also makes possible overall reversal of film. 

The advantages evident in practice are summarized : one-man op- 
eration; ease of film loading and reel transport; cleanliness; solution 


economy; low film tension, with no stretching or damage; uniformity 
of results independently of the footage; no undue aerial or chemical 
fog caused by partially exhausted or polluted solutions; clean, ener- 
getic development with straight H&D curves; and flexibility in use 
or extension to future developments. 

The disadvantages evident are space requirements; drums some- 
what heavy and of a rather high cost; the need of cutting film greater 
in length than 200 feet; and incomplete removal of solution traces 
from the film base without a supplementary bath. 

The 100-foot drum might be eliminated by circulating the devel- 
oper, but the one-turn and the 25-foot drum are both desirable and 

Drums are made of iron with the spiral soldered in place. It was 
found that cold wash water caused condensation and corrosion of the 
drum interior and it was necessary to protect all inside drum surfaces 
thoroughly. The cost might be reduced with wood construction and 
monel staples for film-guides, but special precautions would have 
to be taken to insure a water-tight drum. Trays are made of stain- 
less steel 14 ' 15 ' 19 but could be made of iron and coated with flexible 
air vulcanizing rubber enamel. Because of the great amount of 
soldering, 15 ' 19 stainless steel was not considered practicable for the 
drums, so a hard rubber enamel is used which has proved highly re- 
sistant to solutions, but somewhat susceptible to chipping. 

Acknowledgment is made of the invaluable assistance rendered 
by Emery Huse of the Eastman Kodak Company and to W. J. 
Ellis, W. G. Elder, and L. C. Danby of the Department staff. 


1 "The Davidge Developing Apparatus," J. Soc. Mot. Pict. Eng., XXIV 
(May, 1935), p. 452. 

2 CRABTREE, J. I., AND IVES, C. E. : "Rack Marks and Airbell Markings on 
Motion Picture Film," Trans. Soc. Mot. Pict. Eng., No. 24 (Oct., 1925), p. 95. 

CRABTREE, J. I.: "Uniformity in Photographic Development," J. Soc. Mot. 
Pict. Eng., XXV (Dec., 1935), p. 512. 

DUNDON, M. L., AND CRABTREE, J. I.: "Investigations on Photographic 
Developers Sulfite Fog by Bacteria in Motion Picture Developers," Trans. Soc. 
Mot. Pict. Eng., No. 19 (Sept., 1924), p. 28. 

3 DUNDON, M. L., AND CRABTREE, J. I.: "The Fogging Properties of De*- 
velopers," Trans. Soc. Mot. Pict. Eng., XII, No. 36 (1928), p. 1096. 

4 IVES, C. E.; "A Roller Developing Rack for Continuously Moving the 
Film during Processing by the Rack-and-Tank System," /. Soc, Mot. Pict. Eng. t 
XXIV (Mar., 1935), p. 261*. 

176 R. S. LEONARD 

IVES, C. E.: "An Improved Roller Type Developing Rack with Stationary 
Drive," /. Soc. Mot. Pict., Eng., XXXI (Oct., 1938), p. 393. 

5 CRABTREE, J. I.: "The Development of Motion Picture Films by the Reel- 
and-Tank Systems," Trans. Soc. Mot. Pict. Eng., No. 16 (May, 1923), p. 163. 

* DUNDON, M. L., AND CRABTREE, J. I.: "The Effect of Desensitizers in 
Development," Trans. Soc. Mot. Pict. Eng., No. 26 (1926), p. 111. 

7 FRAYNE, J. G., AND PAGLIARULO, V. : "The Influence of Sprocket Holes upon 
the Development of Adjacent Sound-Track Areas," J. Soc. Mot. Pict. Eng. 
XXVIII (Mar., 1937), p. 235. 

8 CRABTREE, J. I.: "Directional Effects in Continuous Film Processing," 
/. Soc. Mot. Pict. Eng., XVIII (Feb., 1932), p. 207. 

CRABTREE, J. I., AND WADDELL, J. H.: "Directional Effects in Sound Film 
Processing II." J. Soc. Mot. Pict. Eng., XXI (Nov., 1933), p. 351. 

9 MOYSE. H. W., AND WHITE, D. R.: "Borax Developer Characteristics," 
/. Soc. Mot. Pict. Eng., No 38 (May, 1929), p. 445. 

CARLTON, H. C., AND CRABTREE, J. I.: "Some Properties of Fine-Grain 
Developers for Motion Picture Film," Trans. Soc. Mot. Pict. Eng., XIII, No. 
38, (1929). p. 406. 

CRABTREE, J. I., AND IVES, C. E.: "A Replenishing Solution for a Motion 
Picture Positive Film Developer," J. Soc. Mot. Pict. Eng., XV (Dec., 1930) p. 627. 

10 CRABTREE. J. I., PARKER, H., JR., AND RUSSELL, H. D. : "Some Proper- 
ties of Two-Bath Developers for Motion Picture Film," /. Soc. Mot. Pict. Eng., 
XXI (July 1933), p. 21. 

11 RUSSELL, H. D., AND CRABTREE, J. I. : "An Improved Potassium Alum Fixing 
Bath Containing Boric Acid," /. Soc. Mot. Pict. Eng., XXI (Aug., 1933), p. 139. 

12 CRABTREE. J. I., AND IVES, C. E : "Dye Toning with Single Solutions," 
Trans. Soc. Mot. Pict. Eng., No. 36 (Sept., 1928), p. 967. 

CRABTREE, J. I., AND MARSH, W. : "Double Toning of Motion Picture 
Films," /. Soc. Mot. Pict. Eng., XVI (Jan. 1931), p. 57. 

Eastman Kodak Co., "Tinting and Toning of Eastman Positive Motion 
Picture Film." 

13 CRABTREE, J. I., AND MUEHLER L. E.: "Reducing and Intensifying Solu- 
tions for Motion Picture Film," J. Soc. Mot. Pict. Eng., XVII (Dec., 1931), p 1001 

14 MITCHELL, W. M.: "Applications of Stainless Steels to the Motion Picture 
Industry," J. Soc. Mot. Pict. Eng., XXIV (April, 1935), p. 346. 

LAQUE, F. L. : "Iconel as a Material for Photographic Film Processing 
Apparatus," J. Soc. Mot. Pict. Eng., XXIV (April 1935) p. 357. 

LAQUE, F. L.. "Some General Characteristics of Chromium Nickel-Iron 
Alloys as Corrosion-Resisting Materials," /. Soc. Mot. Pict. Eng., XXXII 
(May, 1939), p. 505. 

16 CRABTREE, J. I.. MATTHEWS, G. E., AND Ross, J. F. : "Materials for the 
Construction of Motion Picture Processing Apparatus," J. Soc. Mot. Pict. Eng., 
XVI (March, 1931) p. 330. 

19 Eastman Kodak Co., "Motion Picture Laboratory Practice." Chap. IV, 
Developers and Development. 

17 Ibid. Chap. V, Fixing, Washing, and Drying. 

18 Ibid. Chap. VI, Processing Motion Picture Film.. 

19 Ibid. Chap. VII, Processing Difficulties. 

* Ibid. Chap. XII, Intensification and Reduction. 




Summary. Some warbled frequency films, intended as signal sources for acoustical 
response measurements, appear to have been made and used without full realization of 
the true nature of the warbled signal and the manner in which such a signal is affected 
by a non-linear transmission system. It is pointed out that the warbled signal is a 
frequency-modulated signal; hence the signal may be represented by a carrier fre- 
quency and a series of side-frequencies, all of which are steady and discrete. It is 
pointed out, and substantiated experimentally, that the signal must be regarded in this 
light when considering the effect on it of a non-linear transmission system. The fre- 
quency structure of one "warble film" in use is calculated and shown graphically. 
Fundamental requirements for a suitable warbled frequency film having sinusoidal 
modulation are discussed and values for modulation rate and for modulation depth are 
recommended. The side-frequency array provided by the recommended modulation 
constants is shown in graph form. Expressions are derived giving the frequency 
relationship and relative amplitudes of the side-frequencies resulting from the non- 
sinusoidal frequency modulation which contains two components of modulation rate, 
one component having an associated phase constant. The side-frequency structures 
corresponding to some assumed combinations of two rates are calculated and illustrated. 
Certain assumptions are made for distortion or departure from sinusoid of a modulat- 
ing frequency and the effects on the side-frequency structure are shown. From the 
latter calculation recommendations are derived for tolerances of departure from 
sinusoidal modulation for a warbled frequency film. 

It is well known that the use of steady sinusoidal signals for re- 
sponse measurements on a loud speaker system in an auditorium 
results in a highly irregular response curve; that is, the acoustic 
response varies widely between two frequencies closely spaced. Fig. 
1 illustrates a portion of such a frequency response measurement 
made using steady sinusoidal signals, the microphone being placed 
near the center of a small live room. The data are plotted to an ex- 
panded frequency scale. It will be observed that the characteristic 
consists of a succession of peaks and valleys joined by abrupt slopes. 

* Presented at the 1939 Fall Meeting at New York, N. Y.; received October 7, 

** Altec Service Corp., New York, N. Y. 




(J. S. M. P. E. 

To point out an extreme, a change in response of as much as 24 db in a 
frequency shift of 1 cps or 1 per cent was recorded. 

The irregularities revealed by such measurements are usually 
ascribed to cancellation and reinforcement resulting from phase 
differences between direct and reflected waves reaching the micro- 
phone simultaneously. Although various substitutes for steady 
sinusoidal signals have been proposed, the substitute in widest use 
is the "warble tone." A warbled signal is often thought of by users 
of warbled frequency films as a signal of only one frequency, and 
therefore sinusoidal, but of a frequency which varies in value back 
and forth over a limited range. Its action is accordingly regarded 


FIG. 1. Acoustical response measured in a small room, using sinusoidal 


as preventing the establishment of standing waves by virtue of elimi- 
nating the coincidence of a direct and a reflected wave of the same 
frequency, and as providing an average response over the frequency 
interval of the modulation by integrating the response to an infinite 
number of instantaneous frequencies over the interval. This con- 
cept of the warbled signal will be referred to in the following dis- 
cussion as the "swinging-frequency" viewpoint. 

The warbled signal is an instance of frequency modulation. The 
instantaneous value of the pressure or velocity of the acoustic waves 
produced by sinusoidal frequency modulation are proportional to 

Feb., 1940] 



q = sin (ut + k sin pt) 

where = / the mean frequency, 

, mu> 

K = 

m = per cent modulation expressed as a fraction, 

p, the modulation rate 


Several writers 1>2<3 have shown that eq. 1 may be expressed in the 

q = Jo (k) sin ut -f- .7i(jfe)[sin ( + fit sin ( fit] 
+ / 2 ()[sin (w + 2fit + sin ( -2 fit] 
-j- Ja(/fe)[sin (w 4" SM)^ ~ sin (<o SM)^] 
+ e/c. (2) 



"'O SO SO IOO 110 120 130 

FIG. 2. Predicted transmission of warble signal: 100 cps modulated 
10% ten times per second. Transmission system assumed to have 
characteristic a. The output signal amplitude will vary with location 
of / . (b) Output variation as predicted by "swinging-frequency" 
analysis; (c) output variation predicted by "side-frequency" analysis. 

The coefficients Jn(k} are Bessel functions of the wth order and of 
argument k. 

Here we find that the warble tone is made up of a number of com- 
ponents, each sinusoidal, having frequencies f + p, f p, f + 2p, 
fo 2p, etc. This concept of the warbled signal will be referred to 
in the ensuing discussion as the "side-frequency" viewpoint. 

180 E. S. SEELEY [j. s. M. P. E. 

It is a common error to assume that the two viewpoints are merely 
two ways of saying the same thing and that both lead to the same 
conclusions. On the contrary, the two viewpoints might lead to very 
different predictions of the effect on the signal of* a transmission sys- 
tem possessing frequency discrimination. This point is well illus- 
trated by considering a transmission system having a transmission 
band only a few cycles in width. Such approximately is the char- 
acteristic of a wave analyzer. Assume a signal having a mean fre- 
quency of 100 cps sinusoidally modulated =*= 10 per cent, that is, from 
90 to 110 cps, at a rate of 10 times per second. Assume also, to 
simplify the problem, that the pass-band is two cycles in width as 
shown in Fig. 2(a) and that frequencies outside the pass-band are 
infinitely attenuated. The "swinging-frequency" viewpoint would 
lead to the prediction illustrated in Fig. 2(6) that with the pass-band 
centered at 100 cps the indication would be 6.4 per cent of the steady 
100-cycles reading; that is, the instantaneous frequency would be 
expected to be between 99 and 101 for 6.4 per cent of the time. 
When the pass-band is centered slightly to either side of 100 cycles 
the reading would increase to a maximum of 20.5 per cent when 
centered at 91 or 109 cps; after which it would fall off to zero below 
89 cps and above 111 cps. 

The "side-frequency" mode of analysis leads to quite different pre- 
dictions. The assumed warble signal is found by eq. 2 to be made up 
of components having frequencies and amplitudes as follows : 

Frequency Amplitude Relative to Unmodulated Signal 

100 cps 0.76 

90 and 110 0.44 

80 and 120 0.12 

70 and 130 0.02 

Below 70 and above 130 Negligible 

These figures show that the energy is not spread continuously over 
the frequency range 90 to 110 cps, but that it is lumped at discrete 
frequencies spaced 10 cps apart and that an appreciable part of it is 
found as low as 80 and as high as 120 cps. The prediction one must 
make of the manner in which this composite signal would be trans- 
mitted over our hypothetical system is indicated in Fig. 2(c). 

Predictions (b) and (c) in Fig. 2 are so different as to be irreconcil- 
able and it should not be difficult to establish clearly which is correct. 
Accordingly the described warble tone was applied to a wave analyzer 
having a transmission characteristic approximating that illustrated 


in Fig. 2(o) sufficiently to serve our purpose. The largest reading 
was obtained at 100 cps; no energy was found between the side fre- 
quencies; the readings at 80 and 120 cps were easily obtained, and 
energy indications were obtained at 70 and 130 cps. 

The conclusion is that the side-frequency viewpoint must be em- 
ployed or serious misconceptions may result. To go further, it seems 
safe to say that the frequency-modulated signal, despite the mecha- 
nism by which it is produced, is not a sine wave the frequency of which 
is continuously varying, but a series of sinusoidal waves of definite 
frequencies separated numerically by the modulation rate. In this 
connection, the mathematician might say that a sinusoidal wave of 
varying period is not a sinusoidal wave at all, and therefore that the 
"swinging-frequency" concept is based upon a contradiction of terms. 

The merit of the warbled signal for acoustical measurements resides 
in its multiple-frequency character. However, the side-frequency 
analysis warns us that the number of frequencies of significant ampli- 
tude may be quite small and the separation frequency quite large. 
Each of the components, it will be observed, is itself a steady sinu- 
soidal signal and therefore capable of setting up standing waves and 
susceptible to interference. Appraised in the light of its side- 
frequency composition, a warbled signal might be found inadequate 
even though it may have been judged quite satisfactory from the 
"swinging-frequency" viewpoint. An example is a warble film in 
fairly common use, the modulation of which is given as = t 5 per cent 
and the rate about 8 per second. The frequency composition of the 
signals produced by such modulation is shown for the lower fre- 
quencies in Fig. 3. This illustration leads to the conclusions that the 
patterns for mean frequencies of 300 cps and above are satisfactory 
since they contain a considerable number of components in a reason- 
ably compact arrangement, but that at the lower mean frequencies 
an insufficient number of components exists and the carrier contains 
too large a percentage of the total energy; thus tending too much 
toward the pure sinusoidal signal. 

The following generalization will be found useful in planning a more 
suitable type of modulation : 

Approximately 96 per cent of the energy in the composite signal is contained 
in components having frequencies equal to the mean frequency and all side fre- 
quencies up to and including the th side frequency, where n is equal to the 
modulation index k in eq. 1. (3) 



[J. S. M. P. E. 

From the foregoing rule, which can be verified numerically from a 
table of Bessel functions, and the fact that the separation of the side 
frequencies is equal to the modulation rate p, several relationships 
may be deduced. The "effective band-width," that is, the width 
of the frequency range in which is included approximately 96 per cent 
of the total energy, is equal to the side-frequency separation times 
the number of separations : 

e.b.w. = 2pn = ^Pj^fo = 2mf = 2A/ (4) 


where m is per cent modulation, expressed as a fraction 
/o is the mean frequency 

A/ is the apparent extreme instantaneous deviation from the mean 

> fco TO ao o IPO 

FIG. 3. Spectral distribution of warble signals: modulation = fc 5%; rate 
8 cps; modulation sinusoidal. 

Hence the effective band-width is independent of p. 

Note. It is interesting to observe that the "effective band-width" is equal to 
that which would be expected from the "swinging-frequency" mode of analysis. 
In this particular respect the two viewpoints lead to the same conclusion. 

The number of components included in the "effective band-width" is 

2m/ i i _ e.b.w. 

2n + 1 


+ 1 


It is apparent from these relations that, to obtain a large number 
of frequencies within the band at low mean frequencies, it is necessary 


to employ a large value for m/p, or, for a given band-width, p must be 
small. Since it is desirable to obtain in the effective band as large a 
number of components of different frequency as practicable, reduction 
of p to very small values would be indicated. 

Reduction of warble rate, however, introduces an opposing con- 
sideration in the form of amplitude modulation. The superposition 
of a number of sinusoidal signals differing slightly in frequency would 
rightly be expected to result generally in amplitude variation at low 
rates, the rates being related to the difference in frequencies. The 
relation between amplitudes and phases of the signals set up by fre- 
quency modulation represents a special case of multiple-frequency 
superposition and for this case the composite wave amplitude is 
constant. Any disturbance of this relationship, however, whether 
it be due to frequency-selective action of the direct transmission 
system or due to a selective action of reflecting surfaces or space on 
the reflected waves, introduces an amplitude variation, the lowest 
frequency of which is numerically the frequency separation of the 
signals ; that is, the frequency-modulation rate. If the rate is made 
very low, the needle of the level indicator will tend to follow the 
amplitude variations. There is a distinct disadvantage in indicator 
needle unsteadiness in that greater care is required to determine the 
average reading. The avoidance of serious needle unsteadiness 
leads to the lower limit of modulation rate. The TA-4145 output 
meter probably is used more widely in theaters, in this country and 
elsewhere, than is any other type of level indicator. The ballistic 
properties of this meter are such that when a pulsating (square wave) 
d-c signal varying from zero to three volts 2.5 times per second is ap- 
plied, the needle oscillates ==1.3 db on the 6-volt scale. Hence if 
serious unsteadiness is to be avoided when using this meter, the 
warble rate should not be less than 2.5 cps. 

Using a warble rate of 2.5 cps and selecting 5 as the minimum num- 
ber of frequencies in the effective band centered about 40 cps leads 
to a figure for m of 12.5 per cent. While a larger number of fre- 
quencies would be preferable, they could be obtained only by further 
reduction of the warble rate or increase in m above 12.5 per cent, 
both of which appear undesirable. Fig. 4 illustrates the frequency com- 
position of signals modulated in accordance with these constants for 
several low mean frequencies. It appears undesirable to increase 
the modulation above 12.5 per cent due to the overlapping of the 
band-widths about adjacent mean frequencies. 

184 E. S. SEELEY [j. s. M. P. E. 

Fig. 5 presents an appraisal of the adequacy of these suggested 
warble signals. The response characteristic shown in Fig. 1 is used 
for the purpose. Based on the response as drawn, calculations were 
made of the response to warble signals having a composition like 
that of the proposed 40-cycle warble signal but located at various 
frequencies below 100 cycles. Above 100 cycles the calculations 
were made for a warble signal having a composition like that of the 
proposed 100-cycle warble tone. The band-width of the 40-cycle 
signal being 10 cycles and that of the 100-cycle warble tone being 25 
cycles, the true mean response of the system was calculated for these 
band-widths centered at the same mean frequencies as used in the 





FIG. 4. Spectral distribution of warble signals: modulation ==12.5%; 
rate 2.5 cps; modulation sinusoidal. 

previous calculation. The results of the latter calculation are plotted 
as circles. For the 15 points computed below 100 cycles the average 
departure from the mean (averaged without regard to the sign) was 
0.69 db and the maximum departure from the mean was 1.8 db. 
The improved composition of the signal used above 100 cycles resulted 
in an average departure from the mean of 0.58 db and the maximum 
departure of 1.3 for the 9 points computed. A further calculation 
was made to include in the same comparison the results that would be 
obtained using a signal proposed by Barrow, 4 which he calls the 
"multi-tone." This signal as illustrated consists of components of 
equal amplitude distributed uniformly over the band-width. Since 

Feb., 1940] 



such a signal is also subject to the limitation which controls the lower 
limit of modulation rate for the warble tone, it is necessary to assume 
the same number of components in the multi-tone as are obtained in 
the warble tone. The composite multi-tone and warble tone signals 
are given equal amplitude if the individual components of the multi- 
tone are made equal to one over the square-root of the number of 


FIG. 5. Calculated response using complex signal. Solid lines join points 
measured using sinusoidal signal; circle points: "mean" calculated for 
bands 10 cps wide below 100 cps and 25 cps wide above 100 cps. 

Below 100 cps Above 100 cps 

Multi-tone Warble-tone Multi-tone Warble-tone 

Average dep. 

from "mean" 
Max. dep. 

from "mean" 

0.55 db 0.69 db 0.54 db 

-2.0 db -1.8 db -1.1 db 

0.58 db 
-1.3 db 

components. Such a signal might be expected to yield results some- 
what more accurate than those given by the warble tone. As shown 
in Fig. 5, this was found to be the case, but the difference as revealed 
by this series of computations is so small that the two would be 
considered to yield practically equal accuracy when used for measure- 
ment of acoustic response. 

186 E. S. SEELEY [J. s. M. p. E. 

It has also been pointed out 5 by Barrow that a warble tone, to 
neutralize effectively the standing waves resulting from a single re- 
flecting surface, must have a spectral distribution which embraces 
half a wavelength of the standing wave; that is, from a node to an 
antinode. At 40 cps, a modulation of 12.5 per cent produces a 
band-width of 10 cps, which will embrace node to antinode if the path 
difference is 54 feet or more. If the reflecting surface under con- 
sideration is normal to the path of incident sound to the microphone, 
the surface must be 27 feet from the microphone. If the surface is 
half this distance away only half the node-to-antinodal wavelength 
will be under measurement, with the result that variations theoreti- 
cally as great as 8 db could be obtained by moving the microphone to 
various phase locations in the standing wave if a single reflected wave 
is considered and the surface is assumed to be without absorption. 

No way out of this dilemma can be suggested if it is desired to make 
measurements near the walls of an auditorium. To increase the 
modulation sufficiently to bridge the gap would mean that the 
entire low end of the spectrum would be plotted at a single point. 
However, the picture is not as dark as might be expected since re- 
flected waves reach the microphone from a large number of surfaces, 
none of which is without absorption; and the practice of making 
measurements at a number of points in an auditorium without attach- 
ing too much significance to the readings at any one point also tends 
to mitigate the shortcomings of a reasonable modulation. 

The disagreement of this conclusion with the requirement postu- 
lated by Barrow, viz., that the band-width should extend from node 
to antinode, results from the differing applications considered for the 
warble tone. Barrow's analysis applies to the use of the warble tone 
for reverberation measurements where presumably the interest is 
centered in sound-absorbing materials, whereas the present discussion 
is restricted to the warble tone in response measurements wherein the 
auditorium with its variable local characteristics are the significant 

The foregoing discussion is based on sinusoidal modulation. An 
analysis was made to reveal the effect of departures from true sinu- 
soidal modulation. Formulas giving the side-frequency amplitudes 
were derived for modulation at two simultaneous rates with a phase- 
constant defining their phase relationship. These expressions are 
derived in an Appendix. Three particular cases are illustrated in 
Fig. 6 for mean frequencies of 40 and 100 cps: a 25-per cent second 

Feb., 1940] 



harmonic of the modulating rate with no phase constant, a 25-per 
cent second harmonic having a 90-degree displacement from the 
fundamental, and a 50-per cent second harmonic displaced 90 degrees. 
It is interesting to note that distortion in the modulation-time law 
of the kind which makes the half -cycles asymmetrical about the mid- 

f o .toocjs. 

FIG. 6. Effect of departures from sinusoidal modulation. Vertical bars 
represent spectral distribution of sinusoidal modulation of ==12.5% at 2.5 
cps. Circles represent spectral distribution for modulation illustrated and 
denned by 

/ = /[! + mi sin 2a- Pit + m 2 sin (2* p t t + 6)] 
For all cases illustrated mi = 0.125; p\ = 2.5 cps; pi 5 cps. 

points produces asymmetry of the side-frequency array. Conversely, 
distortion which leaves the half-cycles symmetrical about their mid- 
points does not disturb the symmetry of the side-frequency ampli- 
tudes about the mean frequency but does alter the amplitudes. Al- 

188 E. S. SEELEY [j. s. M. P. E. 

though not shown, third-harmonic distortion produces qualitatively 
similar results. 

In the asymmetrical cases it will be observed that the non-sinu- 
soidal modulation contains a single component that may tend to 
dominate the composite signal and for that reason it would appear 
desirable to limit the second harmonic to 10 or 15 per cent of the 
fundamental for the condition of zero phase-angle. In the sym- 
metrical cases it is found that a 25-per cent second harmonic gives 
rise to a spectral distribution which is no less satisfactory than that 
corresponding to sinusoidal modulation. A 50-per cent second har- 
monic with zero phase-angle may, however, produce undesirable 
accentuation of a small number of the components, and also increases 
the amplitude of the more remote side frequencies. 

Appropriate restraint forbids the drawing of broader generaliza- 
tions from study of the few particular cases discussed. The method 
of analysis afforded by equations 13, 13(o), and 13(6) of the Appendix 
is submitted as providing a rigorous evaluation of the spectral dis- 
tribution when the frequency-modulation time-law is completely 
expressed by two rates and a phase constant, or a practicable ap- 
proximation when one harmonic of the fundamental warble rate pre- 

Conclusions. No implication is intended here that the warble 
tone is the best type of signal for acoustical response measurements. 
A recent revival of interest in this field of measurement and the 
presence of a number of warble films in the field constitutes the 
reason for presenting this discussion of warble tone properties. 

It is the writer's conclusion that any predictions of the effect of a 
warble tone are fraught with danger unless they are based on its 
multiple-signal aspect. This side-frequency viewpoint indicates that 
warbled signals for use in acoustical response measurements should 
have, particularly at low mean frequencies, as large a modulation as 
permitted by the desired separation of the mean -frequency values, 
and that the warble rate should be as low as allowed by the ballistics 
of the indicator to be used. The criterion of standing wavelength for 
determining modulation of a warble tone for acoustical response 
measurements is regarded as impracticable, due to the ambiguity of 
the problem when an effort is made to consider multiple reflections 
and multiple reflecting surfaces. Warbled signals based on a modu- 
lation of =*=12.5 per cent and a rate of 2.5 per second are believed 
satisfactory down to at least 40 cps, and the multi-tone seems to 


offer little promise of significant improvement over such a signal. 
In any practical case, frequency modulation will be non-sinusoidal 
to some degree, and the several common types of distortion of the 
modulating rate analyzed indicate that 10 to 15 per cent of any 
anticipated types of distortion and larger amounts of particular types 
of distortion may be tolerated without significant impairment of the 

Analysis of Non-Sinusoidal Frequency Modulation 

To limit the complexity of the final expressions, modulation at only two simul- 
taneous rates related through a phase constant 8 will be considered. 
The signal is defined by 

q = sin [ut + ki sin n'it + ki sin (ntt + 8)] (/) 

where m and m are the two modulation angular velocities and ki and k% the cor- 
responding modulation indices, equal to miu/m and m^w/nt, respectively, m\ and 
m t being the modulation factors. 

Expansion of the last term in the brackets of eq. 1 results in the form 

q = sin [ut + ki sin nil + (ki cos 8) sin nil + (ki sin 8) sin nil] (2) 

The parentheses are used in eq. 2 since ki and are not time variables and 
(ki cos 0) and (ki sin 0) may therefore, and will hereafter, be treated as constant 

From eq. 2, q is equal to the sine of the sum of four angles and as such it may be 
expanded into the following form : 

5 = [sin w/]]cps (ki sin /ii/)][cps (A sin nit) cos (B cos nil) sin (A sin nit) sin (B cos nil)] 

sin w/] [sin (ki sin /ii<)][sin (A sin nit) cos (B cos nit) + cos (A sin nit) sin (B cos nil)] ,,. 

+ cos ut] [sin (ki sin ftiOHcos (A sin nit) cos (B cos nil) sin (A sin nit) sin (B cos nit)] 

+ cos b>(j[cos (ki sin piO ] [sin (A sin nit) cos (B cos nit) -4- cos (A sin nit) sin (B cos nil)] 

where" A = ki cos 6 and B = ki sin 0. 

To facilitate future reference, eq. 3 is rewritten as follows: 

sin w<][cps (ki sin MiOHI II] 

sin <i)f][sin (ki sin fiiOHIII + IV] 

cos wtUsin (1 sin jiiOlfl II] 

cosw/][cos (ki sin /iiOHUI + IV] (4) 

The sine and cosine functions of the expression x sin <p or x cos <p appear re- 
peatedly in eq. 3. From any standard work on Bessel functions the following may 
be verified: 

cos (x sin <f) - Jo(x) + 2Ji(x) cos 2<p + 2J t (^) cos 4 V + 

sin (* sin <p) = 2Ji(x) sin <p + 2J 3 (x) sin 3> + 2Ji(x) sin 5> 4- (5) 

By substituting I - ^ J for <p in eq. 4 one may obtain 
\ 2 / 

cos (* cos if) J(x) 2Jt(x) cos 2v + 2y(Ar) cos 4y> .... 

sin (x cos if) " 2Ji(x) cos v> 2/j(x) cos 3cp + 2Ji(x) cos 5> . . . . (tf) 

190 E. S. SEELEY [j. s. M. P. E. 

The coefficients Jn(x) are Bessel functions of the first kind, of order n and argu- 
ment *. To substitute eq. 5 and 6 into eq. 5, x must be given three values: 
ki, (k t cos 9), and (& 2 sin 6). To simplify the resulting expressions, the following 
symbols will be employed: 

Kn - Jn(ki) 

Mn = Jn(ki cos 6) 

Nn - Jn(kt sin 0) (7) 

/ in eq. 4 is expanded through the use of (5) and (6) as follows: 

/ - [Mt + 2Mt cos 2ntt + 2M4 cos 4// + ---- ][Nt 2Nt cos 2mt + 2Nt cos 4/ 
- NtNt 2MtNt cos 2/ttt + 2MtN t cos 4/< + ---- 
+ 2MiN t cos 2/ttt 4MtNt cos 1 2^1 + 4MiN 4 cos 2fitt cos 4j< - ---- 
cos 4M2 4MtNt cos 4M2< cos 2Mi< + 4M4#4 cos* 4/uf .... 


- Oo + 2Oi cos 2/2< + 2O cos 4 M 2< + ---- 
where Oo = AfoM - 2A/iM + 2M t N t - ____ 

Oi - -MtNt + Jtotfo + MtNt - MtNt - MtNt + MtNt + MtNt - ____ 

O4 - MtNt + MtNt - MtNt - MtNt - MtNt + M t N + MN 4 - ---- 

O. - -MN, + M,N* + MtN* - M t Nt + ____ etc. (8) 

In a similar manner, it may be shown that 

/// = 2Oi sin MI< + 2Oi sin 3^.1 + 2O sin Butt + ____ 

where Oi - AfiAT + MiNt - MtNt - M,N t + M t N t + Mttft + .... 

O, = -MiNt - M t N 4 + MsATo + MiNt - M f Ni + ____ 

Ot = MiNt + MiNt - MtNt + AfjATo + ____ 

Or - -Mitfe + M>Nt - MiNt + ____ etc. (9) 

II = 2Pi sin 2/tit + 2Pt sin 4/2< + 2P sin G/ttt + . . . . 

where Pj = MiN t + MiN, + MiNi - MiNt - M t N t + ____ 

Pi = -MiNi - MiN t + M,Ni + MiNi + .... 

Pt - MiNi - MtNt + M t Ni + ____ 

P> - M,N t - MtN, + ____ 

etc. (JO) 

IV - 2Pi cos M2< + 2P cos Sittt + 2Ps cos 5ntt + ____ 

where Pi = MtNi + MtNi - MiNi - M t N t + M t N t + M t N t + ____ 

P, -- MtNt + MiM + MiNt + MtNi - MtNt + ____ 

Pt = MoN t - MtNi + MtNi + MtNi + ____ 

Pi - - MtNi + MtNt - MtNt + MtNi + ____ 

etc. (11) 

[One may add indefinitely to the foregoing expressions by observing certain rules : 

(1) The sum or the difference of the M and N order must equal the or the 
P order, as the case may be. 

(2) Only odd orders or only even orders of M and N are involved in a given 
expression, and inspection of eq. 5, 9, 10, and 11 will give the key. 

(5) All possible combinations of M and N orders consistent with the foregoing 
rules are included. 

(4) All M orders have the positive sign; the N*s have the same signs as the 
7's of the same order in expressions 6. The signs of all terms in eq. 8 and 11 
are determined from this rule. The signs of the terms in eq. 9 and 10 result 
from the product of three factors: (a) (6) (c). (a) = 1 as determined by the 
rule just expressed for eq. 8 and 11. (b) = 1 if the difference between the N and 
M order is required to obtain the associated or P order; otherwise it is +1. 
(c) = 1 if the M order is higher than the N order; otherwise it is +1. 

Employing expressions 8, 9, 10, and 11 as well as 5 and 6, the expansion of 3 or 
4 into trigonometric series may now be completed. The first step in the expan- 
sion results in terms consisting of products of sines or cosines or both of two or 
three angles. These may be finally expanded through the following trigono- 


metric relationships. The right-hand side of each equation is a short-hand 
representation of the expression appearing between the two equality symbols.] 

4 sin A sin B sin C = -sin (A + B + C) + sin (A - B + C) + sin (A + B - C) - 

sin (A B - C) = y^j: sin (A =*= B * C) 
4 cos A sin B sin C = -cos (A + B + C) + cos (A - B + C) + cos (A + B - C) - 

cos (A - B - C) = ^t cos (A == B * C) 
4 sin A cos B sin C = -cos (A + B + C) - cos (A - B + C) + cos (A + B - C) + 

cos (A - B - C) = y^T cos (A =t B C 
4 cos A cos B sin C = +sin (A + B + C) + sin (A - B + C) - sin (A + B - C) - 

sin (A - B - C) = j^l sin (A =t= B C 
4 sin A sin B cos C = -cos (A + 5 + C) + cos (A - B + C) - cos (A + B - C) + 

cos (A - B - C) = J] cos (A B =*. C) 
4 cos A sin B cos C = sin (A + B + C) - sin (A - B + C) + sin (A + B - C) - 

sin (A - B - C) = ]T? sin (A =*= B =*= C) 
4 sin A cos B cos C = sin (A + B + C) -f sin (A - B + C) + sin (A + B - C) + 

sin (A - B - C) = ^| sin (A =t B * C) 
4 cos A cos B cos C = cos (A + B + C) + cos (A - B + C) + cos (A + B - C) + 

cos (A - B - C) <= J^| cos (A =t B == C) 

2 sin A sin B = -cos (A + B) -f cos (A B) = 2 1 cos (A B) 
2 cos A sin B = sin (A 4- B) sin (A B) = S sin (A B) 
2 sin A cos B = sin (A + B) + sin (A - B) = 2J sin (A B) 
2 cos A cos B = cos (A + B) + cos (A - B) = St cos (A =*= B) (12) 

The final expansion results in a sine series and a cosine series: 

tf = ?1 + 2 (13) 

gi = KoOo sin o>/ + KoOiSl sin (w /)< + KoOz2t sin (o> =t 2/2)/ + . . . 

t sin (a >ii)< + KiOi^l sin (<o =*= M i =*=/)' + KiO 2 ^? sin (wm2iu)l + . . . 

sin (a) 2 M i)< +[K20i]^ sin (w 2 W =*=/)< +IX 2 O2^| sin ( 2/n 2/ii)< + . . . 
sin (o> 3/)< +IK 3 Oir sin (o> 3/ M2)' + KsOzl sin (<a 3/^1 2/)/ + . . . 


cos ( M2)< + KoPz2 cos (o> 2 M 2)< + XoPaSt cos (w 3/)/ + . . . 

cos ^ st/ * 1 **&+** Pi COS ^" M1 2 M2)<+JK:iP3 T cos(w 

4- ... (13b) 

[Terms may be added indefinitely to the two-fold infinite series (13a) and (I3b) 
by observing that 

(1) the coefficient of ^2 increases by 1 per term as progress is made to the right; 
the coefficient of j*2 increases by 1 per term as progress is made downward; and 

(2) where four + or signs follow the sigma, it will be observed that they 
appear in two arrangements in a given line or a given column. These two ar- 
rangements alternate across the page and downward. ] 

In the solution of a problem, terms must be evaluated until K n P n or K n O m 
become negligible in value. 

When = 0, N = 1 and all other N orders are zero. Hence from eq. 10 and 
11 all orders of P are zero, and J 2 0. 

When 6 = -, M = 1 and all other M orders are zero. Hence the odd-order 

192 E. S. SEELEY 

O's and the even-order P's become zero, and expressions 13a and 13b are accord- 
ingly simplified and the series converge more rapidly. 

Each term in eq. 13a and 13b represents four frequencies: ( + an\ + bm), 

(w am + but), etc. In some cases two or all four of these may be equal. 

When a harmonic relationship exists between MI and m, a considerable number of 
terms from eq. 13a or 13b or both may enter into the evaluation of the amplitude 
at each frequency. If q^ is not equal to zero, the total amplitude at each fre- 
quency is made up of a number of sine components from 13a and a number of 
cosine components from 13b. In this case the modulus at the frequency in point 

c /(sum of contributions) *\ , /(sum of contributions) *\ 1/! (14a) 

* " V from 13a ) + V from 13b ) 

The phase-angle of the component having the frequency in point is 

, /sum of contributions from 13a\ 

On - tan" 1 { -. r~r ) (14b) 

\suni of contributions from 13b / 

A property of eq. 13a and 13b that proves valuable in providing a check on 
computations is the fact that the sum of the squares of the total amplitudes at 
each frequency is equal to 1. This is equivalent to observing that the modulus 
of q is 1, which follows from its definition 1. 


1 VAN DER POL, B.: "Frequency Modulation," Proc. I. R. E., 18 (July, 1930), 
p. 1194. 

1 RODER, H. : "Amplitude, Phase, and Frequency Modulation," Proc. I. R. E., 
19 (Dec., 1931), p. 2145. 

3 SHEA, T. E., MAC.NAIR, W. A., and SUBRIZI, V. : "Flutter in Sound Records," 
J. Soc. Mot. Pict. Eng., XXV (Nov., 1935), p. 403. 

BARROW, W. L.: "The Multitone," /. Acoust. Soc. of Amer., 10 (April. 
1939), p. 275. 

BARROW, W. L.: "On Interference Elimination with The Warble Tone," 
J. Acoust. Soc. of Amer.. 3 (1932), p. 562. 



Summary. The motion picture is a product of science. There is ample his- 
torical material available for those who wish to convince themselves of this fact, but a 
brief review is given of the work of Muybridge and Marey in order to clarify the cause 
of their inventions. The ensuing discussion centers around the question, "Has 
science maintained its interest in the motion picture and has it utilized its advantages 
to its full extent?" 

In this paper the word "science" is taken broadly and includes research, dissemina- 
tion of knowledge, and industrial application. Motion picture's application to 
science is divided into two distinct categories and are discussed in detail: 

(1) The motion picture as an aid to scientific research; 

(2) The motion picture as a medium for the dissemination of knowledge. 

The paper concludes uith descriptions and demonstrations of interesting ma- 
terial from the files of the Rolab Photo-Science Laboratories. Also an inside view 
is given of production activities of an unusual character. 

As we all know, the motion picture is a product of science. The 
names of Muybridge and Marey are familiar to us and we do not 
need to go into detailed discussions and biographies. There is ample 
material available for those who wish to obtain more information 
about the origin and history of the motion picture. However, let us 
refresh our memory for a moment in order to examine the actual cause 
of these early inventions of 50 years ago. 

Necessity is the mother of invention, as we have been told. We 
might add here human curiosity as a possible incentive of invention. 
This certainly was the case with Muybridge, the photographer, who 
wished to find out whether or not a galloping horse leaves the ground 
momentarily. We know the way in which he solved this problem, 
as well as other problems which came up in the field of animal and 
human locomotion, not realizing at the time that the battery of 24 
cameras which he used would give rise to the motion picture industry 
as we know it today. 

Professor Marey, the physiologist, also tried to solve his scientific 
problems of men and animate and inanimate matter, with the aid of 

* Presented at the 1939 Fall Meeting at New York, N. Y. 
** Rolab Photo-Science Laboratories, Sandy Hook, Conn. 




U. S. M. p. E. 

photography. However, in place of Muybridge's 24 cameras he 
employed a single camera with a rotating shutter producing a single 
negative with multiple exposures of the object moving across the 
field. Later Marey devised a photographic gun to study birds in 
flight, in which the mechanical principles of the modern motion pic- 
ture camera were incorporated. 

The question may now be asked, has science maintained its interest 
in the motion picture and has it utilized its advantages to its full extent ? 
The answer is "NO!" For approximately thirty years science, which 

FIG. 1. Taking motion pictures of a carbon particle on the point of a 
needle. (/) Camera, (2) objective, (5) carbon particle on needle, (4) (5) 
light-sources, (6) back drop, (7) rotating device. 

here includes education, did not avail itself of the motion picture. 
Only within the last fifteen to twenty years have scientists and edu- 
cators begun to use motion pictures again, with reluctance to be sure, 
which today has not been overcome. 

Let us now review the advantages that the motion picture has to 
offer in the field of science, "science" to be understood in its broadest 
meaning. For the sake of clarity let us divide the application of 
motion pictures into two categories: 

(A) The motion picture as an aid to scientific research. 

(B) The motion picture as a medium for the dissemination of knowledge. 

Feb., 1940] 



(A) Let us consider the first item. It has been realized since its 
invention that the motion picture has made us masters over the 
elements of time. We can do with time as we can do with space with 
the aid of telescopes and microscopes. In other words, the motion 
picture is a tool that permits us to investigate phenomena invisible to 
the unaided eye, because they may be too fast or too slow or they 

FIG. 2. Complete microcinematographic apparatus. (1) Camera 
timer with motor, (2) camera, (3) observation piece, (4) focusing 
device, (5) microscope, (6) incubator, (7) incandescent lamp, () 
arc lamp, (9) electric exposure meter. 

may also be outside of the visible region of the spectrum. In class 
A above we have at our disposal : 

(1) High-speed motion pictures (slow-motion), 

(2) Normal-speed motion pictures, 

(3) Time-lapse motion pictures, 

(4) Selective spectra motion pictures, such as ultraviolet, infrared, x-ray, etc. 

190 H. ROGER |j. s. M. p. E. 

(B) Now we come to the second item, the motion picture as a 
medium for the dissemination of knowledge. Motion pictures falling 
into this category are obviously entirely different from the ones just 
discussed, although it may be stated that much of the material belong- 
ing in the first category may well be used in the second. 

The status of informative motion pictures is indeed very complex. 
Judging from the many lists and catalogues in circulation there 
seems to be already an enormous wealth of material available. How- 
ever a closer study of this material may lead one to the conviction 
that a great majority of the films are lacking in quality, editorial as 
well as photographic. They are therefore of little value. On the 
other hand, there are a number of excellent films in circulation that 
should be taken as examples by those who produce such films. 

It is not the purpose of this paper to analyze in detail the merits and 
faults of the available motion pictures. This should be done rather 
by those who use them, and it is to be wished that constructive criti- 
cism were given more freely and so made available to the producers. 

Let us now examine the agencies where scientific motion pictures 
are made. Among them are 

Scientific institutions, laboratories, universities, museums, associations 

Motion picture concerns with special departments 

Governmental departments, Federal and State 

Industrial concerns making films mainly to advertise their products 

Manufacturers of motion picture equipment 

Private groups 

Individual scientists and educators 

The interests and purposes of these groups differ widely; hence the 
difference in type and quality of pictures produced by them is very 
great, indeed. For this reason an impartial board of review whose 
business it would be to evaluate all films submitted and to put their 
stamps of approval on films of real scientific value, would serve a good 
purpose and at the same time would save a great deal of time and 
effort for the users of films who have to select their films from lists, 
and do not know their quality. Such a board would be of help in 
raising the standard of film quality. In spite of the present achieve- 
ments, we may still consider the production of scientific films to be 
in its beginning, and much work remains to be done to fulfill the need 
for a fairly complete library. Furthermore, very little has been done 
so far in the field of advanced learning. We are in need of material 
giving more exhaustive information on special subjects, such as are 


available in scientific books. Because of the fact that too many 
films are general and elementary in character the motion picture is 
not as yet a matter of importance to quite a number of scientists. 

I should like now to discuss some of the more technical phases and 
to give here an inside view into some production activities of an 
unusual character, taken from the files of the Rolab Laboratories. 

The general routine of production in the scientific field is the same 
as in the entertainment field. However, the script is very likely 
subject to constant changes and modifications, brought about by the 
outcome of experimentation. Here is a typical example of the 
amount of work that might be required. A script calls for a scene 
in which germs are shown being killed by a germicide. Manufac- 
turers of chemicals, tooth paste, mouth washes, foods, refrigerators, 
containers, etc., always seem to have this subject in mind. This 
appears to them a very simple matter. One has only to put under 
the microscope a few of the many billions of bacteria which are sup- 
posed to be menacing humanity on all fronts. Next we add the 
germicide, one of the many that are widely advertised and obtainable 
in any drugstore. Now one should see that spectacular phenome- 
non the destruction of the germs. 

Here is a summary of the actual procedure : 

It is assumed that the operator has a good knowledge and prac- 
tical experience in bacteriological technic. It is also assumed that he 
has at his disposal an array of standard bacteriological equipment 
glassware, incubator, platinum-loops, Bunsen burner, sterilizer, 
etc. Instead of starting and isolating his own bacterial cultures, 
which would require a great deal of extra time, he would purchase a 
culture from a type culture collection. Then he would prepare the 
culture medium best suited for this particular experiment, and also 
for good photography, and start subcultures in order to have fresh 
material at hand for each new experiment. He now prepares a 
typical micro slide for observing the bacteria under the microscope. 
A few words about bacteria might be of interest. Many persons seem 
to believe that bacteria when seen through a microscope look like 
lice or worms crawling about. The fact is that most of them do not 
move at all, are extremely small, and are hardly visible at the highest 
magnification. Even experts often can not tell whether they are 
alive or dead. Their shapes are simple little rods, dots, spirals, 
chains. Motile bacteria, such as typhoid, may move, however, 
very rapidly. 



LJ. S. M. P. E. 

Continuing his experiments the operator now tries to add very 
carefully, with a small pipette, a minute drop of the germicide, what- 
ever it may be. We assume that he has found a group of bacteria 
that show up very clearly. To his dismay, the operator will find upon 
adding the germicidal fluid his whole field blurred, the bacteria gone. 
The use of high-powered objectives, with their extremely small 
depth of focus, measured in microns, leaves little chance of success 
for this experiment. The slightest touch or even a change in tem- 
perature will cause a change in focus. 

FIG. 3. Apparatus set-up for taking motion pictures of microscopic 
phenomena at very low temperatures. (/) Camera timer with motor, (2) 
camera, (5) microscope, (4) freezing chamber on microscope, (.->) freezing 
apparatus, (6) light -source 

This work might go on for days or weeks, until a way may perhaps 
be found to show the desired scene, by means of a new technic with a 
moist chamber in conjunction with micromanipulation. However, 
to take motion pictures we have not only to control the action with 
regard to time, but also to take precautions with regard to light and 

The operator proceeds with his photographic problems, and we 


assume that he has his complete microcinematographic apparatus 
carefully adjusted and ready, with his microscope inside an incubator 
regulated for 37.5C. We shall not describe here the various parts 
that have to be manipulated. The operator makes his exposure 
tests with the aid of a test specimen and the exposed strips of film, 
having been developed in the darkroom, are then examined with a 
strong magnifying glass. 

Again we assume that all went well and that the intense light as 
well as the heat, reduced by proper filters, has done no damage to the 
specimen. It may be mentioned that all microorganisms prefer 
darkness for best conditions. 

Skill, perseverence, and good luck may finally produce the desired 
result in the form of a certain length of film which will later be a part 
of the completed motion picture. This is not all fiction: A motion 
picture actually exists of this phenomenon under the title, Action of 
Bacteriophage upon Bacterium Coli. This film was made about 10 
years ago with the collaboration of Dr. Bronfenbrenner at the Rocke- 
feller Institute. 

Taking another typical example, scenes might be required showing 
a living insect. Most of such shots are made at close range so that 
the image on the film is either slightly reduced in size, of natural size, 
or enlarged up to 10 times, depending upon the details desired. 
Many persons believe that this kind of photography is much simpler 
than working through a microscope at 1000X magnification. Those 
who have had experience will believe otherwise. Some say that 
this work is even more difficult. Let us consider some of the prob- 
lems: Most microscopes are designed for high magnification. The 
size of the object is limited, and may not be larger than the aperture 
diameter of the substage condenser. Most objects for the micro- 
scope are practically transparent. On the other hand, small objects, 
such as insects, being opaque, require surface illumination. Therefore 
the small spotlights used for this purpose have to be arranged simi- 
larly to those in a studio, although within a very narrow space. 

Although microscopes may be used for this type of work and there 
is some special equipment available, flexibility is lacking when we 
come to photographing living objects. For years our laboratory has 
paid special attention to this kind of work, known as "photomacrog- 
raphy," and we have developed an equipment consisting of about 
150 parts which may be arranged to suit various purposes and taken 
apart after the job is completed. Here again, as in all close work, the 

200 H. ROGER |j. s. M. P. E. 

lack of depth of focus presents difficulties. Let us again take the 
insect as an example. In order to get all parts sharply into focus we 
may have to work with objective apertures of //32 or even //64. 
This requires a very intense illumination, considering the short ex- 
posures necessary (about VM sec.), and if not cooled down with heat 
absorbers, the heat so created would burn the insect within a second. 

Another point of consideration is the elimination of vibration. In 
order to obtain sharp negatives the object must be placed upon a 
rigid stand or holder. The lens should have no mechanical connec- 
tion with the camera which, having moving parts in its interior, 
would otherwise cause vibrations in the objective. 

To obtain the utmost in sharpness our laboratory has a special 
macrophotographic studio with a floor of poured concrete 13 by 33 by 
3 feet thick, upon which the equipment stands. The building is 
located 1 l /z miles from the nearest traffic. 

In all this type of work, dealing with living material, we encounter 
a variety of problems. Our tiny actors on the stage of the micro- 
scope are often many times as temperamental and as difficult to 
handle as those in Hollywood. In contrast with still photomicrog- 
raphy, which has become a matter of routine in many laboratories, 
micro motion pictures almost always have to deal with living or at 
least moving objects which, considering their size and delicacy, re- 
quire extreme care. The slightest change in environment, light, 
heat, composition of medium, radiations, shock, bacterial infection, 
all have to be taken into account. The cultivation of living tissue, 
for example, offers many such problems. It has even been found here 
that certain types of glass, such as ordinarily used for preparations, 
have toxic effects. The composition of the glass in this case becomes 
almost as important as in the construction of photographic lenses. 

A motion picture we made some time ago of budding yeast, starting 
with a single cell which developed into hundreds of cells, caused us 
much labor. The script called simply for a scene 50 to 60 feet long. 
On various tests we found that it would take six to eight hours for a 
yeast cell to develop into a group filling the entire field. We also 
made preliminary experiments with regard to a proper medium in 
which the cells would stay in focus and not move or shift. Further 
tests were made with regard to light and heat. Then we began to 
take pictures. 

The trouble with the yeast cells is that they do not show whether 
they are alive or dead. After hours of shooting, frame after frame, 


we often had to give up further work because the cells had died. 
The following day we had to start a new culture again. After many 
such attempts we finally succeeded in obtaining a fine record of the 
budding process, having by that time found the correct combination 
of light filters which did not interfere with the growth and the proper 
photographic material. 

One may think perhaps that lifeless material may not lend itself 
for motion picture subjects; yet we have done a great deal of motion 
picture work with inanimate matter. For an industrial picture deal- 
ing with carbon we have taken scenes demonstrating dry flocculation 
of carbon particles as well as the agglutination of particles in liquid 
suspensions. We have balanced a single carbon particle upon a pin- 
point, and have taken a picture of it while the pin was rotated. This 
required a rotating device built with precision, as the slightest inac- 
curacy would have moved the object out of the field at such high 

For several films on colloid chemistry we have been very successful 
in obtaining good film records of the Brownian movement of ultra- 
microscopic particles. These particles, smaller than the wavelength 
of light, can not be seen in the most powerful microscopes except by 
indirect illumination, the principle of the ultra-microscope. This 
principle may be compared with a beam of sunlight entering a dark- 
ened room through a narrow opening. What we actually see are the 
reflections of dust particles floating in the air, which in ordinary light 
would be invisible. We have made other pictures of colloidal gold, 
silver, copper, arsenic, manganese, and sulfur, and also the carbon 
particles suspended in india ink ; also colloidal particles floating in the 
fluids of the human eye. We have succeeded in taking motion pic- 
tures of single particles of cigaret smoke floating in air. They look 
like small globules. For this purpose we constructed a small cham- 
ber that fitted into the microscope, through which smoke was blown 
through small openings. 

To show the precipitation of rubber latex we injected a small 
amount of acid into a drop of latex by means of a micro pipette 
operated with a micromanipulator. The pictures of this action are 
quite spectacular. Other experiments, recorded in motion pic- 
tures, were the making of colloidal silver with the electric arc, the 
cataphoresis of colloidal particles showing their migration in the 
electric field, the process of coagulation, the swelling of glue and 
gelatin, etc. 

202 H. ROGER [j. s. M. P. E. 

Recently we were called upon to take motion pictures of various 
objects subjected to high as well as to low temperatures. For the 
high temperatures we built a miniature furnace having a window 
through which we photographed various kinds of borax crystals 
and other substances. Pictures were taken of a variety of objects 
subjected to gradually diminishing temperatures. 

In a patent infringement case dealing with the manufacture of 
ice cream confections we had to demonstrate various phenomena 
taking place in the molds at various degrees. We obtained beautiful 
pictures of the formation of ice crystals and also capillary actions in 
wood and paraffin. 

Several reels of film were projected in court and accepted as evi- 
dence, which in itself may be considered significant for the future 
use of motion pictures for legal purposes. 

The equipment used for this work had, of course, to be specially 
constructed in such a way as to permit careful control ot the tem- 
perature. It consisted of a freezing chamber through which a re- 
frigerant was circulated, and a cooling unit, a liquid-air con- 
tainer with a copper coil inside. The chamber was fitted to the 
microscope with an opening for the light to enter from below and 
another opening for the objective to enter from above. The prepa- 
ration had to be completely enclosed in order to prevent conden- 
sation of moisture on the surfaces of the optical system. Naturally, 
depending upon the temperature desired, we employed various re- 
frigerants and cold-producing agents such as compresssed carbon 
dioxide snow, usually known as "dry ice." 

A similar set-up was used to show the formation of wax crystals 
in various motor oils for a motion picture for one of the largest oil 
concerns in the country. The picture was made to demonstrate 
what happens in a drop of motor oil when subjected to intense cold. 
In one type of oil we saw large elongated crystals forming inter- 
locking networks, causing the oil to become solid or semisolid, thus 
preventing the motor from starting. In another type of oil the 
crystals remained small and isolated from one another at the same 
low temperature, proving that this oil remains liquid and permits 
easy starting. For these experiments we had to use polarized light. 
A Nicol prism representing the polarizer was fitted into the freezing 
chamber and another Nicol prism acting as analyzer was placed in 
the microscope tube immediately above the objective. 

One interesting problem had to do with focusing. First, we had to 


use high magnification with extremely small depth of focus; second, 
we had to focus upon a plane in the specimen that was invisible at 
room temperature but at which the crystals were to appear quite 
suddenly in the form of very small dots when the freezing point was 
reached. This problem was solved by making the system rever- 
sible. We focused on the crystals as soon as they appeared and 
went back to room temperature to start taking pictures. However, 
we had to make allowances for expansion and contraction in the 
microscope set-up. 

Further descriptions of unusual motion pictures could be given, 
but enough has been said for the present purpose. 

It is gratifying to me not only to witness the increase of interest in 
the use of motion pictures in science including industry and edu- 
cation but also to have taken part in this development. When I 
began my work at the Rockefeller Institute after the World War, the 
making of motion pictures was considered by most scientific workers 
as a hobby, certainly not much respected as a scientific tool. How- 
ever, as soon as films of living cells made their appearance at staff 
meetings, showing the structure and behavior of these cells in a real- 
istic way as had never been possible before, the disregard turned into 
enthusiasm. This was especially true at the time when we could 
claim discoveries, made through motion pictures, in the structure 
of white blood corpuscles, such as a presence of a membrane sur- 
rounding each cell. 

The films of living cells have been demonstrated at many scien- 
tific meetings, here and abroad; however, for reasons not quite 
understandable the Institute has not permitted the films to be copied 
and distributed for the benefit of schools and universities. It is to 
be regretted that many requests for demonstrations have always 
been turned down. 

No attempt has been made in this paper to cover the entire field 
of science. To do so would fill many volumes. There are hundreds 
of laboratories today in which motion picture cameras have found 
uses to a greater or less extent. I have not discussed the use of the 
motion picture in practical medicine and surgery, a field which to- 
day, thanks also to the availability of 16-mm equipment and color 
stock, has found wide applications; 1 nor have I treated here the 
use and the accomplishments of high-speed motion pictures 2 (or so- 
called slow -motion pictures), or of astro-photography. 

Interesting discoveries are in store when the electron microscope is 

204 H. ROGER [j. s. M. P. E. 

put to use and perhaps hooked up with the motion picture camera. 

I can not leave these technical discussions without mentioning a 
few amusing inquiries I have received from time to time. A manu- 
facturer of tooth paste wished to show in motion pictures the germ- 
killing action of his product. After showing him what germs looked 
like through a microscope at high magnification, he insisted that I 
use paramecia or slipper animals, which live only in ponds and lakes, 
and which some schoolboy may have shown him through a magni- 
fying glass. These seemed to explode when a drop of soap solution 
was added. 

The subject of paramecia reminds me of a film I saw about Leeu- 
wenhoek, the inventor of the microscope, who lived about 250 years 
ago. To demonstrate the terrible germs, menacing humanity, 
Leeuwenhoek suddenly turned toward one of the guests in his room, 
made him open his mouth, took an instrument the size and shape of 
a screw driver, and manipulated it as a dentist would use a forceps 
when extracting a tooth. The guest was terrified when the germs 
were pulled out of his mouth "without anesthesia." Then Leeu- 
wenhoek examined with his simple microscope his find on the end of 
his instrument- lap-dissolve "Paramecia or Slipper Animal." 

Another tooth-paste manufacturer, who added some colloidal metal 
to his product, requested me to photograph a tooth having a cavity, 
and at the same time to show the metal particles bombarding the 
germs and food particles. A manufacturer of face preparations 
added colloidal gold to face cream and wished to show the Brownian 
movement of the gold, which he had somewhere heard about. The 
particles at the same time should enter the pores of the skin and 
remove the dirt collected therein. Many requests for photographing 
atoms and molecules had likewise to be turned down. 

The purpose of this paper is to call attention to and to emphasize 
the wider use of motion pictures in fields of science in which they may 
serve a real purpose, aside from merely entertaining someone. As we 
have seen in the beginning, science has a good right to claim motion 
pictures as its own. Naturally, it is not to be expected that all film- 
producing agencies will be able to make good films requiring a great 
deal of technical experience and specialized equipment. Indeed, 
there exists a lack of skilled men who are motion picture experts as 
well as scientists. To fill this vacancy the Rolab Laboratories 
were established for the purpose of assisting and collaborating with 
others in the production of scientific motion pictures by utilizing 


their experience and elaborate and unique equipment, and so save 
time and expense to those with less or no experience and equipment. 
Plans are also under way to train individuals in the making of 
scientific films. 


1 ROGER, H.: "New Uses of Sound Motion Pictures in Medical Instruction," 
J. Soc. Mot. Pict. Eng., XXXII (May, 1939), p. 527. 

'ROGER, H.: "A New Camera Timer for Time-Lapse Cinematography," 
Ibid., XXXII (May, 1939), p. 549. 



Summary. The problems confronting the scientist who inaugurates the unique 
task of preserving in film for the people of the 80th century a complete picture of our 
life in America today; the problem of the life of film, and of its relationship to an- 
cient papyrus that has come down to us over sixty centuries; the method of preserving 
it; the microfilming and preparation of the records; the making of a duplicate film 
on metal; and the entire scope of the project is set forth and discussed. 

What would you do if you were confronted with the problem of 
preserving for the people of the world six thousand years from now a 
complete picture of our life, civilization, and culture of today? This 
question impressed me forcibly when I read some two years ago that 
Dr. Thornwell Jacobs, President of Oglethorpe University near At- 
lanta, Georgia, proposed to do this very thing. The idea struck me 
as being both unique and practicable if it was not just a publicity 
stunt, for in my experience as a motion picture cameraman during my 
travels to all the strange places of Asia and North Africa, I had often 
stood and pondered in some ancient and forgotten city and wondered 
why there was no record of its people and their life. During my life 
I have visited the photographed Chichen Itza, Petra, Baalbek 
Anuradhrapura, Amber, Golconda, Angor Wat, Karnak, and many 
another desolation of stone that had once been a great city. I 
thought of Tut- Ankh- Amen 's tomb and the great store of material 
contained therein, a store of treasure regarding the life of the king 
and of his stately splendor, but nothing of his people, their lives, their 
thoughts, and their knowledge. So, when I saw the article in the 
Scientific American announcing the idea of the Crypt and asking for 
suggestions, I sat down and from my experience as an archaeologist, 
wrote Dr. Jacobs what I thought ought to be put into the Crypt. I 
wrote a twelve-page letter based upon both my experience as a mo- 

* Presented at the 1939 Spring Meeting at Hollywood, Calif.; received 
March 17, 1939. 

** Oglethorpe University, Ga. 


tion picture technician and as a student of the past. Dr. Jacobs in- 
vited me to come to Oglethorpe and talk the project over with him, 
and as a result of my visit, asked me to take charge of the collection 
and preservation of the materials to be placed in the Crypt. So then, 
from merely speculative contemplation of the project I was confronted 
with the myriad details of the actual work. This, however, did not 
present anything that a motion picture technician should be afraid of, 
for we are all called upon every day to do or create miracles in our 
daily work, and as my motion picture experience had been preceded 
by a thorough grounding in physics and science generally, I welcomed 
the opportunity. So that is how I happen to be now doing the work 
of an archaeologist of the eightieth century, anticipating his every 
desire for knowledge in regard to life in America today. With this 
introduction to the history of the Crypt of Civilization, as it is 
called, let us see what technical problems are involved in carrying it 

The first issue involved was that of finding out what kinds of ma- 
terials would survive the march of the centuries -sixty of them 
that would elapse between the closing of the Crypt and its opening 
in the year 8113 A.D., a period as far in the future as our first recorded 
date in the past. Going back to the past and taking my own ex- 
perience and the work of every scientist who could throw light upon 
the subject, I endeavored to find what had survived the period of 
time contemplated. The most helpful work was that of the scientist- 
archaeologist, Lucas, whose analysis of objects taken from ancient 
Egyptian tombs was both practical and scholarly, Then followed the 
making of a list of objects and substances that have come down to 
us comparatively intact without any conscious effort at preservation. 
This list includes stone, clay objects, wood, bone, glue, leather, linen, 
rag paper, copper, gold, silver, pitch, and glass, as well as minor ma- 
terials. So you see we have quite a list of possible materials to work 
with. As practically all paper used in books and newspapers today 
is made from woodpulp, which is not stable, and, in fact, will not last 
a century, this of course must be ruled out. Then how were we to 
preserve all the knowledge contained in books? Knowing that our 
new cellulose acetate film had a life span equivalent to that of rag 
paper, and knowing that the ancient Egyptian papyri were nothing 
more than rag paper and that one of them, the Papyrus of Nu, is now 
nearly 4000 years old, it was but a step in reasoning to assume that 
cellulose acetate if properly prepared and finished would, under the 

208 T. K. PETERS [J. S. M. P. E. 

scientific method we should adopt, be preserved in splendid shape. 
Then, too, we had a metal on our list that would remain fairly stable 
(copper) and by combining this with a greater percentage of nickel 
so that a white metal resulted, from tests made we found that this 
would undoubtedly last the sixty centuries and come through un- 
scathed. So we decided to make two sets of records, one on metal 
and one on cellulose acetate. The cellulose film record is being made 
on regular 35-mm duping positive, which is given an extended washing 
after fixing, then neutralized and washed again. When thoroughly 
dry, it is put into a vacuum machine and treated with the "vaporate" 
process which coats the gelatine emulsion and penetrates it with a 
hard yet flexible coat of varnish. Photographing upon this film with 
a microfilm camera of my own design, utilizing images double the 
motion picture frame size, we are making a record of one thousand 
works covering the entire essential knowledge of the world in every 
branch. Each of the books is an authority on some particular field of 
study and embraces all that we know of science, art, religion, philoso- 
phy, sociology, useful arts, philology, and general works such as the 
Encyclopedia Britannica, Compton's World Book Encyclopedia, dic- 
tionaries of all modern and some ancient languages, and specific en- 
cyclopedias on various subjects such as photography, costumes, auto- 
mobile engineering, medicine, radio, etc. These when photographed, 
printed, and processed are put into 100-ft rolls, and eight of these 
are placed in a glass cylinder closed at one end and with spacers 
of glass between each roll of film. When the cylinder is full it is 
sealed except for a small tubulation extending from the sealed 
end. This is inserted into the manifold of a vacuum pump and the 
air in the cylinder is exhausted to 3 microns, after which helium is 
allowed to enter and the tubulation is then sealed off, leaving the film 
in an atmosphere of helium containing sufficient moisture to keep it 
in a flexible condition during the time it is sealed up. This process 
finished, the glass cylinder is slipped into an asbestos transite cylinder, 
which in turn is sealed off at both ends and then this is enclosed in a 
stainless steel cylinder with a positive closure which is soldered up. 
Preserved in this manner from contact with the air, and protected 
from fire, moisture, cold, or insects, the film will come through intact. 
The metal film is prepared by a combination of new and old processes. 
Some years ago I became interested in the use of metal film but 
quickly realized that its use in theaters, while advisable on account of 
fire hazard, was not practicable inasmuch as it would have neces- 


sitated making radical changes in the projection equipment to obtain 
a result that never could equal projection through the film. For this 
reason it would never be adopted by the industry in general. As a 
means of preserving historical records, however, it has a definite place 
and when I became associated with the work at Oglethorpe Univer- 
sity, I revived my original metal film procedure and prepared to make 
the necessary records on metal. To do this I have evolved an original 
formula, the only one in the process, but one that is essential to the 
success of metal film for historical preservation. The formula in 
question is for sensitizing the metal with an almost instantaneous 
sensitizer allowing the printing of motion pictures on long strips of 35- 
mm metal as rapidly as it can be done by the ordinary photographic 
process used. The difference between this and the ordinary emul- 
sion, however, is in the fact that when my emulsion has been exposed 
to light it becomes hardened and will act as a resist, so that the por- 
tions that are affected by light can be etched by acid to any depth 
required. It has been possible to do this, of course, for a long time 
in fact, every halftone cut is made by this process. But printing on 
metal up to the time of my discovery took several minutes for each 
exposure, hence would be of no value in making the hundreds of ex- 
posures necessary to make even a roll of 100 ft of film. Hence, all 
processes for the use of metal film have used an ordinary photographic 
emulsion coated on the metal, printing the image in the ordinary way, 
fixing, and washing. This, of course, would be subject to all the 
faults of stripping, etc. In my process, the portions not affected by 
light the image, in other words are left in the raw metal, and this 
is etched to a depth of 0.0002 inch and in the place so etched new metal 
of a contrasting color is deposited. This may be oxidized silver, black 
nickel, or black platinum. The amount used is so negligible that cost 
is not a factor. When the deposit is finished, the resist is removed 
by a chemical solvent, leaving the image in black on the white surface 
of the base nickel, and in a form that can not be washed off or eradi- 
cated except by abrading the metal itself. It can be further protected 
by a coating of cellulose acetate or one of the methacrylates and 
should last ten thousand years when protected in the same manner as 
the cellulose film, namely, in cylinders from which the air has been 
excluded and replaced by helium. 

By either or both of the methods described, of course, the complete 
history of the United States which I have accumulated in con- 
temporary photographs will be recorded. In motion pictures we shall 

210 T. K. PETERS (J. s. M. P. E. 

have a pretty complete history since 1897, beginning with the inaugu- 
ration of President McKinley, and covering every salient feature of 
history as it occurred. As the metal film will also take the sound- 
track, from which the sound can be reprojected by reflection by 
means of a special projector I have devised, the voices and images 
of all our great men will be immutably preserved for the people of 
the future. Motion pictures are in both 16-mm and 35-mm sizes, and 
are preserved in a special container having globular ends to withstand 
the pressure when the container is evacuated. Otherwise they will 
be encased in both the transite and stainless steel containers in the 
same manner as the microfilmed records. The entire microfilmed 
record and motion picture deposit will take up only about 250 cubic- 
feet of space in the Crypt out of a total space of 2000 cubic-feet. 
Thus there will be ample room for the many other interesting deposits 
that will go to compose the entire picture. There is a section devoted 
to plastics, those wonderful substances, man made, which have only 
come into existence during the last twenty-five years. In this section, 
which has been accumulated through the courtesy of Dr. Baekeland 
and his staff, are all kinds of objects in use in our daily life, such as 
an electric razor, an electric iron, a toaster, a radio, an electric clock, 
and many other interesting developments made from bakelite, cata- 
linite, tenite, vinylite, plexiglas, micarta, and other combinations of 
synthetic resins so extensively used today. Samples of cloth of vari- 
ous kinds from the finest to the coarsest, and plain, printed, and em- 
broidered, are sealed in glass containers also. Little manikins dressed 
in characteristic costumes of men and women of this day will also be 
placed in glass containers. Models of all our great inventions such as 
locomotives and cars, refrigerators, printing machines, tabulating 
machines, the cotton gin, automobiles, aeroplanes, and things of 
this nature made in miniature form and to exact scale will be included 
as well as blueprints from which they may be reconstructed in case 
civilization has gone back to primitive times. Many actual articles 
which from their ephemeral nature would undoubtedly be lost will 
also find a place in the Crypt. These will include a lady's handbag, 
containing all the gadgets that the normal bag holds, such as lipstick, 
compact, keys, hairpins, etc., as well as many other objects in common 
use. Artificial aids to hearing and eyesight, an artificial skull, artificial 
eyes, arms, legs, and teeth will give our descendants in the one 
hundred and eighty-seventh generation some idea of the method by 
which we repaired natural defects or losses. In order that all the mo- 


tion pictures and other objects contained in the Crypt may be made 
available, projectors for sound and silent film, as well as by reflection, 
for the metal film, will be placed in with the film; and so that the 
many instruments and utensils can have the proper kind of current, 
a wind-driven generator made of permalloy will be on hand ready for 
use when the Crypt is opened. In case the English language is no 
longer in use, a special machine has been devised by me so that a 
knowledge of the English language may be achieved. The basis 
of the machine is an old mutoscope. To this is attached a phono- 
graph by means of shafting and a heavy flywheel. On the shaft 
is fastened a small generator which delivers 12 volts, supplying cur- 
rent for a small lamp just inside the machine. On the top of the 
machine is a strip picture showing clearly that to use the machine all 
you have to do is to turn the crank. On looking into the machine a 
motion picture is shown of a man holding up various objects and doing 
various things; for example, he holds up an apple and says "apple," 
and underneath the object is printed the word, while the sound comes 
out of the speaker. You see a pair of legs running and the word 
"running" is pronounced and spelled for you, and so on. In this way 
1500 basic English words prepared by the Orthological Institute of 
London are made possible without effort spent in deciphering and 
learning them. In addition, of course, the pronunciation is given, for 
even if English is common in that day, the sound will be probably 
as different as the sound of the English of Chaucer now is to us. 
We are preparing two sets of plates, one for Asia and one for the 
European and Western countries. On each of these will appear the 
complete story of the Crypt, what it will contain, and its exact 
location according to the Coast and Geodetic Survey and in rela- 
tionship to Stone Mountain not far away. The Asian tablet will bear 
upon its surface the story in Chinese, Japanese, Maharatta, Telugu, 
Hindu, Urdu, and Sanscrit. The tablet for Europe and America will 
tell the story in English, French, Spanish, Portuguese, Italian, Russian, 
and Scandinavian so that one of these tablets will be sure to be found 
and point the way to the treasurehouse of knowledge. These tablets 
will be sent to libraries, universities, and temples in the different 
countries in the hope that they will be preserved. In addition to this, 
several thousand tickets of admission to the Crypt are being printed 
on metal to be handed down from father to son until the time for 
the opening of the Crypt, so that the memory of the Crypt will be 
kept alive. 


During the Conventions of the Society, symposiums on new motion picture appara- 
tus are held in which various manufacturers of equipment describe and demonstrate 
their new products and developments. Some of this equipment is described in the 
following pages; the remainder will be published in subsequent issues of the Journal. 


During the 1938 Spring Meeting of the Society a group of Simplex 4-Star 
theater reproducing equipments were described. 1 These systems were designed 
to provide high-quality sound for small as well as for large theaters. A group of 
systems was necessary due to commercial considerations requiring the provision 
of most economical combinations consistent with theater sizes. 

The design requirements, based on considerable research, experience, and con- 
sulation, were divided into four groups, as shown in Fig. I. 1 The groups are: 
Sound Mechanism Equipment, Control Equipment, Power Amplifier Equipment, 
and Loud Speaker Equipment. 

There are certain requirements shown here which may be considered as con- 
stants. For the sound mechanism they include the reproduction of 35-mm film 
at 90 feet per minute; adaptability to standard or push-pull reproduction; the 
requirement that normal speed be attained in 2 to 3 seconds; and that the total 
flutter be not greater than 0.15 per cent. The control equipment in each case 
will be required to offer the same number of change-overs with the standard 2000- 
ft reels, with a volume control at each machine. The power amplifier should 
have not more than 1 per cent total harmonic at 50 cycles, with noise level not 
exceeding 35 db. A two-way loud speaker system with multicellular horn may 
also be considered constants of design, each system having the same crossover 
frequency for economical reasons. 

In view of the above requirements it was clearly indicated that for all systems 
the sound mechanism and the control equipment should be identical. The only 
variables of design were the power output of the amplifiers and, consequently, the 
power-handling capacity and coverage of the loud speakers. 

Thus three systems were designed: The small, or A system, for houses up to 
1000 seats; the medium, or B system, for houses up to 2000 seats; and the large, 

* Presented at the 1939 Fall Meeting at New York, N. Y.; received Novem- 
ber 4, 1939. 

** International Projector Corp., New York, N. Y. 




or C system, for houses up to 4000 seats. This manner of division was made in 
accordance with the recommendations of the Research Council of the Academy 
of Motion Picture Arts & Sciences. 8 

The main objective of insuring uniformly good reproduction of sound was at- 
tained in these systems through the careful consideration of design requirements, 
regardless of the size of theater. 

Approximately 84 per cent of the theaters of the country have under 1000 seats, 
and of this total a very large percentage have less than 800 seats. For such a 
large group of houses, therefore, it seems justifiable that every effort should be 




(1000 Seats) 


(2000 Seats) 



(4000 Seats) 

Sound Mechanism 


Power Amplifier 

Loud Speaker 

35-Mm film 
90 Ft/minute 
Standard, push-pull 
Dual channel 

2000-Ft reels 
Volume control at 
each machine 

1% Total harmonic 
at 50 cycles 
35 Db noise level 

Two-way system 
400-Cycle cross- 
Multicellular horn 

2-3 Sec pick-up 
0.15% total flutter 




Power, Coverage 









(15 watt) 




(30 watt) 






(60 watt) 


FIG. 1. Analysis of design requirements. 

made to provide a combination of equipment capable of good-quality reproduction 
without penalizing the exhibitor to the extent of purchasing equipment suitable 
for much larger theaters. With this in mind another system was designed espe- 
cially for houses of less than 800 seats where the Simplex 4-Star A System can not 
be justified for economic reasons. 

In analyzing the requirements for the new system we maintain that the same 
constants of design adhered to in previous equipments must be retained in order 
that there may be no sacrifice in the quality of reproduction. This immediately 
requires that no change be made in the sound mechanism as described above. 
Also, adequate control facilities must be retained since the duties to be performed 
remain the same. Thus the design variable is again the amount of power re- 



It has been determined that an amplifier system capable of delivering 10 elec- 
trical watts affords adequate power for an 800-seat house and provides the margin 
of power required for the reproduction of the increased volume range of present - 

FIG. 2. Volume control amplifier. 

day recordings. Since the cost of manufacturing a 10-watt amplifier is not a 
great deal less than that of a 15- watt unit, this alone will not result in an appreci- 
able saving. Since we do not wish to sacrifice end-results, the economy must be 

PIG. 3. Main amplifier in cabinet. 

made in the elimination of certain facilities which are, of course, useful in opera- 
tion and servicing but are not essential to the actual quality of reproduction. 

First consideration of the facilities which might be sacrificed was given to the 
control, or front-wall, equipment. The use of high-impedance coupling between 

Feb., 1940] 



the sound mechanism and preamplifier, having proved highly successful in pre- 
vious designs, was maintained in the new system. A single front- wall unit is 
provided into which the high-impedance lines from the sound mechanism termi- 
nate. Fig. 2 shows a view of this amplifier in its cabinet. It is a two-stage resis- 
tance-coupled unit having a gain of approximately 45 db. All connections to the 
amplifier terminate on a bakelite terminal board located near the front of the 
cabinet. Screw connections are used here in order to facilitate removal of the 
amplifier chassis for service or replacement. A potentiometer balancing arrange- 

FIG. 4. Stage loud speaker equipment. 

ment for the photoelectric cell voltage is arranged on the terminal board whereby 
the output of the two machines may be equalized. The volume control is located 
on the side of the cabinet to which an extension shaft is coupled providing a means 
of level adjustment at the operating side of either machine. A telephone jack is 
provided for the connection of non-synchronous attachments. Insertion of a 
plug into this jack disconnects the film input so that normal threading operations 
can be carried on without disturbances in the non-synchronous input. 

Sound change-over is accomplished by means of exciter-lamp switching. A 
pair of interlocking switches is provided whereby the operating exciter-lamp volt- 
age is transferred, and a standby voltage applied to the "OFF" exciter-lamp to 
prevent delay in change-over time. 



Fig. 3 shows a view of the main amplifier cabinet with the amplifier installed. 
Resistance coupling is used between this unit and the front- wall amplifier since, 
as in former systems, we have shunned the use of audio transformers in order to 
avoid hum pick-up. The power amplifier is capable of delivering 10 watts of 
audio power with less than 1 per cent total harmonic distortion. Adjustable in- 
verse feedback is employed for the purpose of providing equalization of the speech 
circuit in addition to its normal functions of noise suppression and reduction of 
distortion. Adjustment of the feedback circuit can be readily made by means of 
straps on the terminal board, which is conveniently located on the front of the 
amplifier chassis. 

A separate monitor circuit is built into the amplifier, bridging across the out- 
put line, consuming practically no power and providing a gain of about 15 db 
above the stage line. The monitor unit and its control are mounted on the door 


trrtnjB ra SKOSD 

FIG. 5. Electrical characteristics of Simplex Type E system. 

of the cabinet, and connected to the amplifier by means of a polarized socket 
on the amplifier chassis. 

The amplifier chassis can be drawn forward and tilted upward at approxi- 
mately a 45-degree angle in order to facilitate servicing. All connections to the 
amplifier are of the screw type, located on the terminal board back of the 
protecting plate, permitting easy removal of the unit to facilitate service and re- 

The exciter-lamps are operated on alternating current. A high-wattage lamp 
was selected in order to reduce the 120-cycle modulation, since it is known that 
this factor is dependent on the heat storage or rate of cooling of the lamp filament. 
In addition, a 120-cycle tuned circuit is inserted in the front-wall amplifier which 
serves to attenuate the hum level resulting from a-c lamp operation. 

Fig. 4 shows the two-way loud speaker equipment designed for use with this 
new system. The high-frequency loud speaker assembly consists of a multi- 
cellular, exponential horn for wide-angle distribution, insuring uniform balance 

Feb., 1940] 











throughout its frequency range. The high-frequency unit is a permanent-magnet 
dynamic type, employing a metallic diaphragm. The low-frequency speaker is a 
15-inch permanent-magnet dynamic type enclosed within a folded horn of solid 
and sturdy wood construction. Both the low- and high-frequency units are 
capable of handling more than 15 watts of electrical power, thus giving ample 
margin and dependable service. The dividing network is located back stage with 
the speakers. The overall gain of the system is about 100 db, which is sufficient 
for both film reproduction and the operation of non-synchronous equipment. 

The electrical characteristic may be adjusted over a wide range at either the 
low or high frequencies. Fig. 5 indicates the range of equalization available. 
The recommendations of the Academy of Motion Picture Arts and Sciences, as 
well as the variations in acoustical conditions encountered in the field, have been 
regarded in the selection of these response characteristics. 

The simplicity of installation of this equipment can be seen in Fig. 6. The 
single preamplifier, the exciter-lamp supply, and the two change-over switches 
are located on the front wall. The main amplifier cabinet, including the monitor, 
is located on the side or rear wall. Conduit requirements are reduced to a mini- 
mum, and the space required should not tax even the smallest projection room. 

No sacrifice has been made in the quality of the component parts of this system, 
but rather every effort has been made to obtain the most reliable material avail- 
able, servicing and replacement operations have been simplified and standard 
vacuum tubes, which are universally obtainable and inexpensive, are used 
throughout. Thus an economical sound equipment has been developed for the 
small houses affording a quality of sound reproduction comparable to that found 
in the larger theaters. 

(Two recordings were used for the demonstration, one being Raymond Paiges 1 Dark 
Eyes from Hollywood Hotel released by Warner Bros, the latter part of 1937, recorded 
on standard variable-area track. The other was from Metro-Goldwyn- Mayer's 
Babes in Arms recorded on standard variable-density track. We wish to express our 
appreciation to Major N. Levinson, of Warner Bros., and to Mr. Lester Isaac, of 
Loew's Theatres, Inc., who furnished the films.) 


1 FRIEDL, G., JR.: "A New Sound System," /. Soc. Mot. Pict. Eng., XXXI 
(Nov., 1938), p. 511. 

J MILLIARD, J. K.: "Projects of the Committee on Standardization of Theater 
Sound Projection Equipment Characteristics of the Academy of Motion Picture 
Arts & Sciences," /. Soc. Mot. Pict. Eng., XXX (Jan., 1938), p. 81 ; Bull. Academy 
of Motion Picture Arts & Sciences (June 8, 1937). 


MR. CRABTRBE : This demonstration proves that we can do a lot more than 
is being done at present with the film emulsions that are now available. I 
had previously heard this film in a theater and the sound was terrible. I think 
that if we could always hear sound of the same high quality as we have just 
heard, we would be quite satisfied. 


MR. MORGAN: What high-frequency unit do you use in the system? 

MR. BARNETT: A permanent-magnet dynamic unit, metallic diaphragm. 
The horn is multicellular three cells. 

MR. READ: The speaker mentioned that the lamp was operated on a-c and 
that a 120-cycle filter was used. How great was the attenuation? 

MR. BARNETT: The 120-cycle attenuation is about 10 db. 



The Simplex double-film attachment (Fig. 1) described herein is designed for 
use with the Simplex 4-Star sound system where separate picture and sound prints 
are to be run for reviewing purposes in studios or for showing pre-release prints 
in theaters. 

The equipment consists primarily of a large magazine, in which are mounted 
two take-up shafts and one feed-shaft to accommodate three reels. A film chan- 
nel, connecting the projector mechanism directly to this magazine, detours the 
film around the sound mechanism. This avoids congestion, facilitates threading, 
and permits easy observation during operation (Fig. 2). 

The picture print is in the upper magazine on spindle A . It is threaded through 
the projector in the normal manner, but leaves the projector after passing over 
the lower holdback sprocket and goes through the film-channel over a pair of 
guide-rollers to the take-up reel, which is mounted on spindle C. 

The sound-print feed-reel is placed on spindle D in the lower right-hand corner 
of the magazine. The film runs over a pair of guide-rollers to a special gear- 
driven sprocket which feeds it into the sound mechanism scanner. The film 
passes through the sound mechanism in the usual manner and then to the take-up 
reel, which is mounted on spindle B. There is ample clearance in the lower maga- 
zine to permit the use of three 1000-ft reels with 5-inch hubs. 

For ordinary sound and picture projection, where the picture and sound are on 
the same print, the film is threaded through the projector and sound mechanism 
in the same manner as in standard projection equipment. In this case (Fig. 3) 
standard 2000-ft reels with 5-inch hubs can be used in the upper and lower 
magazines. With a 2000-ft reel in the take-up magazine it is necessary only 
to shift the guide-roller a in the lower magazine. The magazine door is double- 
hinged, and for single-film operation only one-half of the door need be opened. 
The clearance, when 2000-ft reels with 5-inch hubs are used, is similar to that in 
the standard 18-inch magazine. 

Special consideration has been given to the importance of insuring smooth film 
passage through the projector, sound mechanism, and double-film attachment. 

* Presented at the 1939 Fall Meeting at New York, N. Y.; received October 
28, 1939. 

** International Projector Corp., New York, N. Y. 



The guide-rollers are equipped with ball-bearings, and have high flanges on both 
sides. All parts coming into contact with the film are sufficiently undercut to 
prevent damage to the sound-track and picture area. An extra sprocket is pro- 
vided to feed the sound-print from the lower magazine into the sound mechanism. 
This makes threading easy, avoids sharp bends of the film, and thus reduces the 

FIG. 1. Simplex double-film attachment. 

possibility of patches coming apart. This sprocket is located in the upper part 
of the film-channel, and is directly geared to the sound mechanism projector 

The entire feed-sprocket assembly is mounted on a hinged bracket which en- 
gages the feed-sprocket gear with the projector drive-gear, and permits adjust- 
ment for proper mesh of teeth without requiring shims (Fig. 4). When combined 
sound and picture prints are projected, this bracket can be raised to disengage the 
two gears. 





Important features of the Simplex double-film attachment are simplicity of 
adaptation to the projection equipment, elimination of fitting and adjustments on 
the job, and the elimination of shims for alignment purposes. The double-film 
attachment is designed for interchangeable use with Simplex E-7 or Super-Sim- 



FIG. 4. Showing mounting of feed-sprocket assembly. 

plex mechanisms. With this attachment, when used on installations with Super- 
Simplex or Simplex 57 pedestals, downward projection angles to 30 degrees and 
upward projection angles to 3 degrees can be accommodated. 

The overall length of projector, sound mechanism, and double-film attachment 
is increased by only 1 Vt inches over that of the standard Simplex equipment with 
1 8-inch magazines. 



MR. RICHARDSON: What is the idea of projecting picture and sound-film sepa- 
rately with one projector? 

MR. PIRNER: For preview purposes. After the preview all the changes neces- 
sary are made, and then the sound and picture are combined on one print. 

DR. GOLDSMITH : Is this attachment adaptable to the standard projector? 


MR. BRADY: What would be the effect of running^two prints exactly alike, 
simultaneously and in perfect synchronism? 

MR. FRIEDL: That is a very interesting question, although it is irrelevant to 
the paper. I believe what you are asking is the effect of superimposing identical 
prints, in the hope of obtaining stereoscopic effects, and, perhaps, stereophonic 
sound. Several times at this meeting mention has been made of three-dimensional 
pictures as being "on the horizon," and as representing a challenge to our techni- 
cians. This attachment, however, is not intended to achieve that result. This is 
only for projecting a picture film and a sound-film simultaneously, for the purpose 
of previewing. One of its major uses is in preview theaters, particularly in Holly- 
wood. Before going to the expense of a final editing the producers want to try 
the picture on the public, and later edit it according to the reaction of the public 
and the comments of the studio executives. Usually the one film has the picture 
only and the other film the sound-track only. Two complete picture and sound- 
track films could be projected, but you would get only the picture off one and only 
the sound-track off the other. 

MR. CRABTREE : In the case of a grainy picture, what would be the effect of 
simultaneously projecting several identical images? 

DR. GOLDSMITH: It would be interesting to know whether graininess of the 
prints would be reduced by superimposed projection of several identical images. 
Theoretically, and on a basis of statistical averages, this process would reduce 
graininess. But since graininess in average focused pictures is a relatively mild 
fault, it would hardly seem worth while to use multiple projection to reduce it. 


The problem of non-intermittent film projection was treated in detail by Tuttle 
and Reid 1 in 1932. Since then a considerable number of various devices for such 
a projection method has been developed. 

Our improvements are based upon the development of a projector designed in 
such manner that optical compensation is effected by means of a rotating glass 
prism placed between the film and the projection lens. This projector is very 

* Presented at the 1939 Fall Meeting at New York; received October 10, 

** New York, N. Y. 



FIG. 1. Plane parallel glass plate penetrated by a light-ray. 

it* if 1 

FIG. 2. The relation between the displacement and the angle of incidence. 

Feb., 1940] 



simple and reliable in use, and is especially suited for synchronous sound repro- 
duction. Its perfection and efficiency depend, of course, upon eliminating the 
optical errors caused by the glass prism or at least upon reducing such errors to a 
degree where they will not sensibly influence the projection. 

Basic optical laws prescribe the dimensions of the rotating prism as well as its 
optical placement with respect to the mechanism. The latter in turn depends 
upon the size of the film frame and upon the glass material of the prism. 

In Fig. 1 is shown a plane parallel glass plate penetrated by a light-ray. This 
light-ray enters the plate at the angle of incidence a, passes through the plate at 
the angle of refraction /3, and emerges from the plate parallel to the incident ray, 

FIG. 3. Position of the rotating prism, two faces vertical to the 
optical axis. 

but displaced by a distance b. As Fig. 1 shows all necessary angular relations, 
the vertical displacement b can be calculated. 

Letting z equal the length of the path of the light-beam through the glass, and 
n the index of refraction: 

b = z sin (a /8) 
D = z cos /3 

D sin (a - /3) b cos 

b = or D = 

cos /3 

sin (a /3) 

The vertical displacement b is a function of the angle of incidence a, the index of 
refraction n, and the thickness of the plate. Fig. 2 is a plot of this function for a 
value of n = 1.52, a value of unity having been assumed for the thickness D. 



FIG. 4. (Upper) Position of the rotating prism, two edges on 

the optical axis. 
FIG. 5. (Lower) Position of the rotating prism, two edges below 

the optical axis. 

Feb., 1940] 



The function is linear up to the angle of 15 degrees. Therefore, a uniformly 
rotating plane parallel plate causes a uniform displacement of the light-ray so long 
as the tilting angle of 15 is not exceeded. 

Applying this condition to a rotating glass prism, at the instant the prism has 
been rotated 15 degrees, the succeeding image must be projected; hence the neces- 
sary number of prism faces can be calculated: 


2 X 15 


FIG. 6. The curved gate of the film. 

A polygonal prism of 12 faces, therefore, allows a linear displacement. Now 
the necessary thickness of the prism can be calculated. Rotation of the prism 
by 15 degrees corresponds to a displacement of half an image, which is 3.81 mm 
in the case of 16-mm film. In Fig. 2, D was assumed equal to unity; hence, if 
B is height of the half-frame on the film and the value of b is taken as 0.0918, then 

or D = 41.5 mm. 

In the following discussion of the optical conditions the influence of the lens 
upon the light-rays has not been considered. The differences caused by the fact 
that the lens does not give strictly parallel rays are relatively so small that they 
may be neglected if a lens with long focal length is used. 


In the projection of film frames two extreme positions of the prism must be 
considered. The position shown in Fig. 3 does not result in any outstanding 
effect. In Fig. 4 two edges of the prism are located on the optical axis. The 
image seen looking through the prism, hereafter termed visible image, is formed by 
the halves of two succeeding film frames. 

The upper half of the visible image is formed by the upper part of the lower 
frame while the lower half is formed by the lower part of the upper frame. In this 
position of the prism the extreme angle of 15 degrees is exactly achieved but not 
exceeded, and the visible image is free from astigmatism. 

Conditions become more unfavorable with further tilting of the prism (Fig. 5). 
The visible image is composed of a major portion of the upper film frame and a 

FIG. 7. The complete projector. 

minor portion of the lower film frame. The position of the prism faces now below 
the optical axis corresponds here with the unfavorable unlinear section of the curve 
of Fig. 2. The larger part of the visible image has good definition, since it corre- 
sponds to the linear section of the curve; the other part, formed by the lower frame, 
shows astigmatic distortions which increase with further tilting of the prism until 
the upper frame enters into the position shown in Fig. 3. 

The effect of exceeding the 15-degree angle appears as varying astigmatism. 
To reduce this effect it would be useful to utilize a prism with 24, at least 16, faces; 
but the dimensions would be larger. 

During the passage of the film frame from one position to the next, two positions 
exist without defects caused by tilting of the prism ; namely, the positions of Figs. 
3 and 4. Immediately before reaching the position of Fig. 3 the upper part of the 
visible image shows astigmatic distortion. These errors can be reduced by black- 


ing out the edges of the prism in order to render them inactive when in the unfa- 
vorable position of Fig. 5. 

Fig. 5 shows another phenomenon: The peripheral speed of the rotating 
prism is higher than the speed of the film. By using glass having a higher index 
of refraction, the diameter of the prism would be reduced, and hence its peripheral 
speed. A certain choice of index of refraction would nearly equalize the peripheral 
speed of the prism and the speed of the film and thereby minimize the defects 
caused by the different speeds, but unfortunately, glass of such an index can not 
be utilized for our purposes. 

The shrinking of film has a marked effect upon the distortion, and the designer 
must consider seriously this important factor. The reduction of the length of the 
film can be compensated by an adjustable roller in a very simple manner, but in a 

FIG. 8. Close-up of projector. 

projector where the size of the film frame influences the entire optical arrange- 
ment, particularly dimensions of the prism, such compensation is not sufficient. 

The frame of the shrunken film has a smaller vertical dimension than the figure 
used for the calculation of the prism and serious confusion of the projected image 
results. The use of an adjustable curved gate (Fig. 6) allows satisfactory com- 
pensation, the curve corresponding to the maximum size of frame and the prism 
being calculated for this size. Altering the curve in accordance with the amount 
of the shrinkage, then, will compensate for differences in the vertical dimension. 

Guiding the film in a curve causes mislocation of the frame with respect to the 
optical system and incorrect definition is the consequence. This can be avoided 
by using lenses of greater depth of field. A slit limiting the vertical aperture of the 
lens used as stop for the projection lens achieves this condition and simultaneously 
reduces the astigmatic errors very materially. 

A further advantage of using this slit is the following: Projecting a film upon a 
screen one sees a central image on the optical axis and secondary images above 


and below caused by the other prism faces out of action. These disturbing secon- 
dary images must be screened out. The necessary diaphragm should be as dis- 
tant as possible from the projection lenses. The smaller the vertical aperture 
of the lens the more effective is the diaphragm. The reduction of light-intensity 
as the consequence of using the diaphragms mentioned above may be tolerated as 
no rotating shutter is utilized in the projector. 

Furthermore we must consider the fact that standard projection lenses are cor- 
rected for light-rays passing in air. It is necessary to correct the optical system for 
spherical aberration considering the glass prism in the beam of light. 

The steadiness of the film frames in relation to the prism faces has been assumed 
to be perfect. Any unsteadiness of this sort causes distortion on the screen. To 
prevent misalignment of the prism faces with respect to the film frames during 
the rotation of the prism, the prism is driven by the film itself, in such a manner 
that the edges of the prism are located in the optical axis as well as on the dividing 
line between the film frames. 

The chromatic errors caused by the fact that the indices for different wave- 
lengths are different are insignificant. The projection of a still picture renders an 
image showing serious distortion but during the rotation of the prism the images of 
the complementary light-rays are superimposed upon each other throughout the 
symmetrical positions of the prism. 

To reduce loss of light the condenser and the projection lamp are arranged to 
concentrate the light-rays in veftical direction. Finally it is necessary to prevent 
mirages caused by total reflection of the prism faces out of action. The concen- 
tration of the light-beam, mentioned above, and the arrangement of the dia- 
phragms may achieve this result. 

A projector designed according to the above-described optical arrangement 
shown in Figs. 7 and 8 gives satisfactory projection and is particularly suitable for 
sound scoring as the absence of the intermittent movement permits inherently 
continuous motion of the sound carrier. 

The authors are at the present time designing a steel tape recorder in collabora- 
tion with Acoustic Consultants, Inc., to be incorporated in the projector. In 
practice, original records will be made an integral part of the photographic record 
and used for reproduction as sound-on-film. The original steel tape sound nega- 
tive may be kept or the record obliterated and the tape used again for original 
sound recording. 


1 TUTTLE, F., AND REID, C. D. : "The Problem of Motion Picture Projection 
from Continuously Moving Film," /. Opt. Soc. Anter., 22 (Feb., 1932), p. 39; 
J. Soc. Mot. Pict, Eng., XX (Jan., 1933), p. 3. 


MR. OFFENHAUSER: At what speed was the film running? It appears to be 
faster than normal. 

MR. EHRENHAFT : Sixty cycles. The machine was designed for 50 cycles. 

MR. OFFENHAUSER: Have you had an opportunity to make studies of com- 
parative flicker of this type of arrangement vs. that of a conventional mechanism ? 


MR. BACK: The prism has 12 faces, and it has been necessary to screen the 
edges, otherwise the astigmatism would be considerable. With a prism of 16 
faces it is not necessary to screen the edges; there is then no flicker at all. 

MR. TOWNSLEY: Is the compensation for shrinkage automatic? If in a reel 
there are two pieces of film of different shrinkages, is there compensation without 
manual adjustment? 

MR. BACK: Not in this projector. The gate must be adjustable to compen- 
sate for shrinkage while the film is running. 

MR. CRABTREE: What are the fundamental advantages, compared with our 
present intermittent type of projector. Both C. Francis Jenkins and A. J. Hoi- 
man have demonstrated optical intermittents, but from a practical standpoint 
they do not seem to have shown any outstanding advantages. 

MR. BACK: The sound reproduction is much simpler when there is no intermit- 
tency in the film movement. The film runs at constant speed, which is much 
better for the perforations and the emulsion. 

MR. EHRENHAFT: Another advantage is that the film indirectly guides the 
prism, so there is no relative movement between the film and the prism. 

MR. CRABTREE : It has always been claimed that the advantage of the non- 
intermittent projector was that it prolonged the life of the film. That is of no 
real value at the present time. Film is not discarded because of torn perforations 
but because of scratches. In other words, the film is strong enough to withstand 
the intermittent. 

Optical intermittents have many disadvantages, including the difficulty of 
keeping the optical system free from dust and dirt, and there is the matter of ex- 

MR. BACK: The expense is less than with other projectors because the sound 
system is much simpler. 

MR. PALMER: We had a number of continuous projectors operating at the 
World's Fair this summer, and films have run through them as many as 1200 
times without any damage to the perforations. The films have to be replaced 
because they become scratched, not because the perforations are damaged. The 
scratching is not usually caused by the projector gate, but by winding the film 
in a continuous roll; the friction of one layer against the other causes the scratches. 

MR. HOLSLAG: Is it necessary to illuminate two full frames? 

MR. BACK: No; only one frame and about one-eighth of the next. 

MR. CRABTREE: I noticed a rather bad flicker at the top edge of the picture. 

MR. BACK: That was caused by blacking out the edges of the prism. The 
prism is a 12-faced prism; when we use a 16-face prism it is not necessary to black 
out the edges, and then we have no flicker at all. 



The editors present for convenient reference a list of articles dealing with subjects 
cognate to motion picture engineering published in a number of selected journals. 
Photostatic copies may be obtained from the Library of Congress, Washington, D. C., 
or from the New York Public Library, New York, N. Y. Micro copies of articles 
in magazines thai are available may be obtained from the bibliofilm Service, Depart- 
ment of Agriculture, Washington, D. C. 

American Cinematographer 

20 (Dec., 1939), No. 12 

Testing New Weston Meter (pp. 533-534, 538) 
Use of Fine Grain Positive Emulsions for Variable Density 

Film Recording (pp. 535-537, 564) 

M-G-M Builds Unique Camera Boom (pp. 539-540, 572) 
Reeves Builds 16-35-Mm. Developer (pp. 541, 566) 
Reproduction of Film Exposed 40 Years Ago (p. 546) 
Smoothing Scene Transitions (pp. 551-554) 

British Journal of Photography 

86 (Nov. 3, 1939), No. 4148 
Progress in Colour (pp. 659-660) 
A Note on the Permanence of Agfacolor (pp. 663-664) 

86 (Nov. 10, 1939), No. 4149 
Progress in Colour (pp. 674-676 

86 (Nov. 17, 1939), No. 4150 
Progress in Colour (pp. 685-686) 


19 (Nov.. 1939), No. 11 
Syncrosound System (pp. 12-13) 
Television Economics. Pt. X (pp. 23-25) 

Electronics and Television and Short- Wave World 

(Nov., 1939) 
American Methods of Film Transmission (pp. 630-633) 


21 (Oct., 1939), No. 10 

Die gerauscharme Atelierkamera der Askania-Werke 
A.-G., Berlin-Friedenau. (Noiseless Askania-Werke 




Studio-Camera) (pp. 235-237) 







Beziehungen zwischen Bild- und Tonsensitometrie. 
(Relation between Image and Sound Sensitometry) 
(pp. 237-241) A. NARATH 

Das Mehrfachkopieren von Kinofilmen. (Multiple Prints 
from Motion Picture Film) (pp. 241-243) O. BENDER 

Philips Technical Review 

4 (Oct., 1939), No. 10 
An Acoustic Spectroscope (pp. 290-291) 
The Efficiency of Loud Speakers (pp. 301-307) 

Photo Technique 

1 (Dec., 1939), No. 7 
A Flexible Time Lapse Outfit (pp. 28-31) 




The History of Photography; Its Relation to Civilization and Practice; Dr. 

Erich Stenger; translation and footnotes by Edward Epstean (1939), Mack 

Printing Co., East on, Pa. 

Here is a book on the history of photography written by the man who, after 
Professor Eder, is the most competent in Germany to write upon such a subject. 
The translation into English, faithfully carried out, is by Mr. Edward Epstean 
of New York, who has done more than anyone in recent years to make photo- 
graphic history available to the English-speaking public. Mr. Epstean has 
given much of his time and money to the preparation of translations of the 
most important French and German works on the subject. 

The present book is interesting from two points of view: (/) the early history 
of photography and the earlier stages in the development of photographic proc- 
esses and their applications are dealt with in great detail, and will be a valuable 
source of reference; (2) the book represents a picture of the present German 
viewpoint of the subject. Mr. Epstean points out in his preface that it is difficult 
to write history objectively, and that any criticism that might be raised as to the 
preference of the author for his Fatherland might apply equally to history books 
written in other countries. A presumably "controlled" book on photographic 
history may in itself, however, be an interesting historical document. We must 
admire Mr. Epstean for arranging for the publication himself when the inter- 
national situation led his original publisher to disavow any sponsorship of the 

There is perhaps a tendency to place too much emphasis on information as to 
the first people who started to do things which later developed into something 
worth while, and not enough information about what they led to. However, 
there are many who will be interested in ready references to the early historical 
material, and to them the book will be valuable. 

Mr. Epstean has added many footnotes, correcting certain misstatements in 
the text, and amplifying some of the material. 






Officers and Committees in Charge 

E. A. WILLIFORD, President 
S. K. WOLF, Past-President 
W. C. KUNZMANN, Convention Vice-President 
J. I. CRABTREE, Editorial Vice-President 
S. HARRIS, Chairman, Papers Committee 
J. HABER, Chairman, Publicity Committee 
H. GRIFFIN, Chairman, Convention Projection 
E. R. GEIB, Chairman, Membership Committee 
H. BLUMBERG, Chairman, Local Arrangements 

Reception and Local Arrangements 






Registration and Information 

W. C. KUNZMANN, Chairman 


Hotel and Transportation 

E. O. WILSCHKE, Chairman 





J. HABER, Chairman 



Convention Projection 

H. GRIFFIN, Chairman 




Officers and Members Projectionists Local 310, IATSE 


236 1940 SPRING CONVENTION [J. s. M. p. E. 

Banquet and Dance 

M. C. BATSEL, Chairman 




Ladies' Reception Committee 

MRS. O. F. NEU, Hostess 

assisted by 




Miss L. A. MOVER, Social Director, Chalfonte-Haddon Hall 


Headquarters. The headquarters of the Convention will be the Chalfonte- 
Haddon Hall, where excellent accommodations have been assured, and a recep- 
tion suite will be provided for the Ladies' Committee. 

Reservations. Early in March room reservation cards will be mailed to mem- 
bers of the Society. These cards should be returned as promptly as possible 
in order to be assured of satisfactory accommodations. 

Hotel Rates. Special rates have been guaranteed by the Chalfonte-Haddon 
Hall to SMPE delegates and their guests. These rates, European plan, will be 
as follows: 

Four Lower 

Floors Ocean View Ocean Front 

Room for one person $ 3.50 $ 4.00 $ 5.00 

Room for two persons 6.00 7.00 8.00 

Parlor Suite, for one 10.00 12.00 14.00 

Parlor Suite, for two 14. 00 16. 00 18. 00 

(All bathrooms at Haddon Hall have hot and cold running fresh and salt 


If American plan rates are desired the hotel room clerk should be advised 
accordingly when registering. An additional charge of $3 per day per person 
will be added to the above-listed European rates for three daily meals, American 
plan. Members and guests registering at the hotel on the American plan will 
pay only $3 for the SMPE banquet scheduled at Haddon Hall on Wednesday 
evening, April 24th. If registered on the American plan, the clerk at registration 
headquarters should be advised accordingly when procuring your banquet 

Parking. Parking accommodations will be available to those who motor to 
the Convention at the Chalfonte-Haddon Hall garage, at the rate of 5Q for day 
parking or $1.25 for twenty-four hours. These rates include pick-up and delivery 
of car. 

Registration. The registration and information headquarters will be located 
at the entrance of the Viking Room on the ballroom floor where the technical and 
business sessions will be held. All members and guests attending the Convention 

Feb., 1940] 1940 SPRING CONVENTION 237 

are expected to register and receive their badges and identification cards required 
for admission to all the sessions of the Convention, as well as to several motion 
picture theaters in the vicinity of the Hotel. 

Technical Sessions 

The technical sessions of the Convention will be held in the Viking Room of 
the Hotel. The Papers Committee plans to have a very attractive program on 
papers and presentations, the details of which will be published in a later issue 
of the JOURNAL. 

Luncheon and Banquet 

The usual informal get-together luncheon will be held in the Benjamin West 
Room of Haddon Hall on Monday, April 22nd, at 12:30 p.m. The forty-sixth 
Semi-Annual Banquet and Dance of the Society will occur on the evening of 
Wednesday, April 24th, in the Rutland Room of Haddon Hall an evening of 
dancing and entertainment for members and guests. 

Ladies' Program 

A specially attractive program for the ladies attending the Convention is being 
arranged by Mrs. O. F. Neu, Hostess and the Ladies' Committee. A suite will 
be provided in the Hotel where the ladies will register and meet for the various 
events upon their program. Further details will be published in a succeeding 
issue of the JOURNAL. 


At the time of registering, passes will be issued to the delegates of the Con- 
vention admitting them to several motion picture theaters in the vicinity of the 
Hotel. The names of the theaters will be announced later. 

Atlantic City's boardwalk along the beach offers a great variety of interests, 
including many attractive shops and places of entertainment. 


Convention Vice-President 



At a meeting held on January 10th at the Engineering Societies Building, New 
York, a paper was presented by J. A. Maurer of the Berndt-Maurer Corporation, 
on the subject of "Motion Picture Production in 16-Mm." 

Recent improvements in film stocks, apparatus, and laboratory technics have 
greatly improved the print quality attainable when photographing pictures di- 
rectly on 16-mm film. With the availability of high-quality 16-mm sound- 
recording equipment, these developments have led to considerable activity in pro- 
fessional direct 16-mm production. 

Mr. Maurer broadly surveyed the equipment, materials, and services existing 
in the 16-mm field, appraised the results attained, and forecast what may be ex- 
pected from the application of still newer materials and technics not now in general 

Results with currently available materials were demonstrated, including the 
reproduction of examples of sound and picture duplicates on Kodachrome, a field 
that has recently assumed considerable commercial importance. 

The meeting was attended by approximately 200 persons and closed with an 
extended discussion of the paper. 


At a meeting held on December 9th at the meeting rooms of the Western Society 
of Engineers, Chicago, a paper by Mr. M. Wenzel of the Wenzel Company, 
Chicago, on the subject of "New Developments in Theater Motion Picture and 
Sound Projectors" was read by Mr. J. S. Scanlon of the same company. The 
paper dealt with recent developments in the design of theater equipment and the 
results attained therewith. The meeting was well attended and a lively discus- 
sion followed the presentation. 


On December llth, at the Bell & Howell Building, Hollywood, the subject of 
"Color 1939" was discussed by L. E. Clark. 

In addition, an exhibition was given on Cinecolor 16-Mm Reduction Prints 
accompanied by an inspection of the Cinecolor reduction process, by A. M. 





Volume XXXIV March, 1940 



Large Size Non-Rotating High-Intensity Carbons and Their 

Application to Motion Picture Projection 


The Development and Practical Application of the Triple-Head 
Background Projector B. HASKIN 252 

Motion Picture Auditorium Lighting B. SCHLANGER 259 

Automatic Slide Projectors for the New York World's Fair. . . . 

F. TUTTLE 265 

The Vocoder Electrical Re-Creation of Speech 

H. DUDLEY 272 

The Objective Measurement of the Graininess of Photographic 
Emulsions A. GOETZ, W. O. GOULD, AND A. DEMBER 279 

Regulations of the National Board of Fire Underwriters for the 
Storage and Handling of Nitrocellulose Motion Picture Film 311 

New Motion Picture Apparatus 

A Flexible Time-Lapse Outfit 


Film Splicer for Developing Machines 


Current Literature 342 

1940 Spring Convention at Atlantic City, N. J., April 22nd to 
25th 344 

Society Announcements 347 





Board of Editors 
J. I. CRABTREE, Chairman 




Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscription or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., East on, Pa., or Hotel Pennsylvania, New York, N. Y. 
Published monthly at East on, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 

West Coast Office, Suite 226, Equitable Bldg., Hollywood, Calif. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1940, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is not 
responsible for statements made by authors. 


* President: E. A. WILLIFORD, 30 East 42nd St., New York, N. Y. 

* Past-President: S. K. WOLF, RKO Building, New York, N. Y. 

* Executive Vice-President: N. LEVINSON, Burbank, Calif. 

** Engineering Vice-President : D. E. HYNDMAN, 350 Madison Ave., New York. 
N. Y. 

* Editorial Vice-President: J. I. CRABTREE, Kodak Park, Rochester, N. Y. 

** Financial Vice-President: A. S. DICKINSON, 28 W. 44th St., New York, N. Y. 

* Convention Vice-President: W. C. KUNZMANN, Box 6087, Cleveland, Ohio. 

* Secretary: J. FRANK, JR., 356 W. 44th St., New York, N. Y. 

* Treasurer: R. O. STROCK, 35-11 35th St., Astoria, Long Island, N. Y. 


* M. C. BATSEL, Front and Market Sts., Camden, N. J. 

* J. A. DUBRAY, 1801 Larchmont Ave., Chicago, 111. 

** A. N. GOLDSMITH, 444 Madison Ave., New York, N. Y. 
** H. GRIFFIN, 90 Gold St., New York, N. Y. 

* P. J. LARSBN, 29 S. Munn Ave., East Orange, N. J. 

* L. L. RYDER, 5451 Marathon St., Hollywood, Calif. 

** A. C. HARDY, Massachusetts Institute of Technology, Cambridge, Mass. 

* H. G. TASKBR, 14065 Valley Vista Blvd., Van Nuys, Calif. 

* Term expires December 31, 1940. 
** Term expires December 31, 1941. 





Summary. The high-intensity, direct-current arc between small, copper-coated 
carbons operated in coaxial alignment without rotation with a reflector optical system 
has achieved a widespread and growing popularity over the past few years for theatri- 
cal projection of motion pictures. This type of light-source has now been extended 
to include larger carbons and higher currents. These larger carbons of this type 
with the proper optical system will give substantially higher light on the motion picture 

Fundamental facts about the arc behavior and the conditions necessary to obtain 
stable and steady operation with these larger carbons are described. The correlation 
of the luminous characteristics of the arc with the optical system is reviewed. The 
performance of a new arc with a suitable optical system is given from the standpoint of 
offering possibilities for projection. Carbon consumption rates, arc current and 
voltage, and light on the screen are discussed. 

Throughout the entire history of motion picture projection there 
has been constant progress in improving the efficiency and quality of 
the projection and the amount of light available on the screen. 
From the beginning, light for projection in motion picture theaters 
has been furnished almost exclusively by the carbon arc because of 
its high intrinsic brilliancy. However, the early carbons, and optical 
systems used to utilize their light, were a far cry from the carbons, 
lamps, and optical systems of today. Some of the more important 
changes are the application of the high-intensity arc to motion picture 
projection, 1 ' 2 making possible the large theaters of today; the use of a 
mirror and small carbons instead of a condenser and large carbons for 
low-intensity arc projection; 3 ' 4 improvements in condenser lenses for 
high-intensity carbons; the advent of the so-called "Hi-Lo" high- 
intensity reflecting arc lamp ; and more recently, the extension of the 
high-intensity arc on an economical basis to the medium and small 

* Presented at the 1939 Fall Meeting at New York, N. Y.; received Novem- 
ber 13, 1939. 

** National Carbon Company, Fostoria, Ohio. 


242 JOY, LOZIER, AND SIMON [j. s. M. p. E. 

theaters by the simultaneous improvement in mirror optical systems, 
and simplification of high-intensity lamp design made possible by the 
development of the small size, non-rotating, copper-coated "Suprex" 
carbon. 5 ' 6 These simpler types of high-intensity reflecting arc 
lamps and "Suprex" carbons have become standard equipment in 
a majority of the medium size theaters. 

The characteristics of this system have been described before; 6 ' 6 
two of its outstanding features are high efficiency and simplicity. 
The ever-present demands for increased screen illumination have led 
to research to extend these high-intensity arcs with non-rotated car- 
bons to larger sizes using the reflector system. The purpose of this 
paper is to present for consideration and to describe the character- 
istics of larger-size non-rotating high-intensity direct-current carbons 
and discuss means of utilizing them in a manner which will make it 
possible to double the screen light obtainable with systems in common 
use today, and to do this in some cases with less energy input into the 
arc. These results represent several years of research and develop- 
ment work on the theoretical and practical aspects of the problem. 


The accepted practice of not rotating the positive carbon and the 
approximately coaxial alignment of the positive and negative car- 
bons have resulted in desirable mechanical simplification of the lamp 
mechanism and the problem of arc control. This arrangement has 
given stable burning with 6-mm, 7-mm, and 8-mm "Suprex" positive 
carbons, the latter of which is used with currents up to 65 amperes. 
Attempts to use a 9-mm positive carbon in the lamps designed for 
"Suprex" carbons met with serious difficulties when the arc current 
reached approximately 75 amperes. At about this current a transi- 
tion was observed in the nature of the arc stream. This transition 
was characterized by the occurrence of a relatively dark "tongue" 
in the arc stream which appeared to originate at the tip of the nega- 
tive carbon and extend some distance into the arc stream. The ap- 
pearance of the arc stream at currents below and above this transi- 
tion point is shown in Fig. 1(A) and (B), respectively. This phe- 
nomenon of the appearance of this tongue in the flame from the 
negative carbon was observed and described some years ago by Bas- 
sett. 7 With the coaxial alignment of electrodes, this tongue from the 
negative carbon is directed into the positive crater and causes the arc 
to become unstable, and the light is erratic and unsteady. At cur- 



rents considerably higher than 75 amperes, this tongue goes over to 
the familiar bright inner "core" or tongue appearing to be attached 
to the negative carbon of standard well known arcs such as the 125- 
ampere arc between the 13.6-mm high-intensity positive carbon and 
Vie-inch "Orotip" negative carbon. In this latter case the negative 
carbon is used at an angle with the positive and the bright tongue 
in the arc streams up in front of the positive crater instead of directly 
into it, and so causes no instability. 


FIG. 1. Sketches showing the appearance of the direct- 
current arc with coaxially aligned electrodes: (A) With 
no negative "tongue"; (B) with negative "tongue." 

Different factors were observed to have some small effect on the 
current at which this instability occurs but in no case was a factor of 
safety obtained which was sufficient to insure reliable operation with 
the large non-rotated carbons in coaxial alignment. However, it 
was observed that very satisfactory operation could be obtained 
without rotation of the positive carbon when the negative carbon was 
placed at an angle with the positive carbon, the range of 30 to 40 
degrees appearing to give best results. Other angles of burning than 

244 JOY, LOZIER, AND SIMON [j. S. M. P. E. 

the range suggested may be made desirable by the use of auxiliary 
magnetic flux or by other design features of the lamp. 

With this arrangement, it is possible to burn large-diameter non- 
rotated high-intensity positive carbons at high currents. In fact, 
it has been possible to burn positive carbons 15 mm in diameter 
above 150 amperes, with indications that still larger carbons and 
higher currents can be used under these same conditions. Operation 
is steady and easily controlled without the use of auxiliary magnetic 


With this knowledge of the proper method of burning these larger 
carbons without rotation of the positive carbon, we proceeded to de- 
velop copper-coated positive carbons in sizes appreciably larger than 
the present 6- to 8-mm "Suprex" carbon range and having desirable 
characteristics for angular-trim reflecting arc lamps. We have also 
included in our development for angular- trim burning, carbons of 
the smaller diameters and have data on positive carbons (with suit- 
able negatives) for this purpose ranging from 5-mm to 15-mm in 
diameter in 1-mm steps. 


Burning Characteristics of 11-Mm Special Non-Rotated Positive Carbons and */- 
Inch Negative Carbons Burned at 35 Angle 

Intrinsic Brilliancy Total 

Approximate Consumption at Center of Crater 
Arc (Inches per Hour) Crater Candles Candle- 
Current Voltage Positive Negative per Sq Mm Power 

110 56 13 2.3 640 30,000 

115 60 16 2.4 700 35,000 

Burning characteristics of an 11-mm copper-coated non-rotating 
positive carbon with a suitable negative carbon are given in Table I. 
This non-rotating angular trim gives steady light and has stable 
operating characteristics at the current and voltage indicated. It 
has been designed to give a higher intrinsic brilliancy than the 8-mm 
"Suprex" at its maximum current of 65 amperes. A comparison of 
the intrinsic brilliancy distribution across the crater face of these two 
carbons is given in Fig. 2. This figure shows also the relative size of 
the light-source from the 11-mm carbon compared with the 8-mm 
"Suprex" carbon. The question of how most efficiently to take ad- 
vantage of this increased brilliancy and larger source diameter in 



motion picture projection brings up the consideration of the proper- 
ties of the optical system. 


The experiences gained with the reflector type lamps used with 
"Suprex" carbons provide much useful data on which to base our 
present considerations. Wide-angle elliptical reflectors having a 

.20 .15 JO .05 O .05 .10 J5 .20 
Radius of Crater in Jnc/ies 

FIG. 2. Comparison of the intrinsic brilliancy and size 
of the crater of the 8-mm "Suprex" positive at 65 amperes 
in a horizontal trim, and the 11-mm carbon at 115 amperes 
in an angular trim. 

collecting angle of approximately 145 degrees are employed with 
"Suprex" carbons, and the magnification ratio of these mirrors is 
about 6.5 to 7.0. The portion of the crater brilliancy curve of the 
8-mm "Suprex," which, when magnified 6.6 times, falls within the 
diagonal of the sound projection film aperture, is shown in Fig. 2. 
Therefore, this shows approximately the portion of the "Suprex" 
crater light which is used in motion picture projection today. It is 
evident that the larger crater of the 11-mm Special carbon will per- 
mit a smaller magnification than 6.6:1 and still obtain good coverage 



[J. S. M. P. E. 

on the aperture. For example, the portion of the curve utilized with 
a magnification ratio of 5.2:1 is shown in Fig. 2. The 145-degree 
6.6:1 elliptical reflector requires an//2.3 projection lens to pass all 
the light which passes through the center of the film aperture. If 
the magnification of the mirror were reduced to 5.2:1 while still 
maintaining a large collecting angle, the relative aperture or "speed" 
of the system would be increased and a projection lens of higher 
"speed" could be filled. 

The various factors affecting the screen illumination, including 
source brightness and speed of the optical system, have been discussed 
previously. 8 ' 9 ' 10 Applying the principles discussed in those articles 
to the data in Fig. 2, it is evident that while the 11 -mm carbon would 

Iterttcro* CA*OK 


FIG. 3. Diagram of the//2.0 reflector arc system. 

give significant increase in screen light with an //2.3 optical system, 
this increase will be more outstanding if used with a faster optical 


Screen Light Measurements on 11- Mm Special Non-Rotated Copper- Coated Carbons 

in Angular Trim 






No Film 





Sides to 



In order to determine experimentally the possibilities of the 11 -mm 
non-rotated copper-coated carbons, an experimental burner was 
assembled. An experimental mirror of proper magnification and 
angular pick-up 11 was obtained from the Bausch & Lomb Optical 



Company. The carbons were not rotated and the negative carbon 
was placed at an angle of 35 degrees from the positive carbon. The 
optical system of this lamp is shown in Fig. 3, where the various 
dimensions of the reflector system are given. A standard sound-film 
projection aperture was used in the system without a film shutter. 
Screen light measurements were taken with a 4-inch focal length 
//2.0 experimental "Cinephor" type projection lens. 11 Weston pho- 
tronic cells equipped with "Viscor" filters were placed at the center, 
sides, and corners of the screen image to measure screen illumination 
and screen distribution. 

The results of these screen light measurements are shown in Table 
II for the same carbons whose burning characteristics are given in 
Table I. 


Screen Light Obtainable with Various Carbons and Optical Systems 

Carbons Amperes 

8-7-mm "Suprex" 65 
13.6-mm H.I. Pos. 125 
Vie" "Orotip" Neg. 
13.6-mm Super 180 

H.I. Pos /* 
"Orotip" Neg. 
11 -mm Special 115 

Non-rotated C.C. 
Pos. 3 /s* Neg. 

Optical System 




No Film Shutter 

Elliptical mirror f/2 . 3-f/2 . 5 6000-8500 
Condenser type //2.S-//2.5 6000-9000 

Condenser type f/2 . 3-//2 . 5 8000-10,000 

//2.0 Mirror de- 
scribed in this 

//2.0 lens 
in this 


In order to understand better the significance of the values given 
in Table II, they should be compared with the amount of screen light 
obtained with conventional lamp systems and carbons in use today 
for motion picture projection. Such figures can be only approximate 
because of the many factors which affect the total luminous flux 
falling on the screen. 

Table III shows the approximate amounts of screen light obtained 
with various well known carbons and optical systems, and that the 
new 11-mm carbon and//2.0 reflector system described in this paper 
will give as much as twice the light of the two high-intensity sys- 
tems most widely used today the 8-mm "Suprex" positive carbon 
with mirror type lamps and the 13.6-mm H.I. positive carbon at 
125 amperes with condenser type lamps. 



[J. S. M. P. E. 

It is of interest to compare the 11 -mm non-rotated positive carbon 
and the 8-mm "Suprex" carbon each with its appropriate optical sys- 
tem, as regards the rapidity with which the screen light and screen dis- 
tribution change as the positive carbon is moved along its axis away 
from the focal point of the mirror. This comparison is shown in 

.15 .10 AS .05 .10 .15 tncntj 
Movement of Positive Carbon a/ong Axis 
Towmrd Mifrr~* \ + A* ay from Mirror 

FIG. 4. Curves showing the change in total screen 
light and the distribution of screen light with movement 
of positive carbon, for the 8-mm "Suprex" and the 1 1-mm 
carbon used with their respective optical systems. 

Fig. 4, and gives an indication of the "flexibility" of the two arcs 
with their respective optical systems. It can be seen that the 1 1-mm 
trim has a better distribution of light on the projection screen. For 
example, at the focal point of the reflector or the position of maxi- 
mum light, the 11-mm carbon with its optical system gives a light 
at the sides equal to 70 per cent of that at the center, whereas the 
"Suprex" carbon gives a light of 63 per cent at the sides compared 
with the center. 


If the positive carbon is moved either toward or away from the 
reflector and the current and relative position of the negative carbon 
with respect to the positive are kept constant, the total light on the 
projection screen will change for these two carbons and optical sys- 
tems according to the two top curves shown in the figure, and the 
distribution will change as shown by the two bottom curves in the 
figure. It is evident that the light falls off much more rapidly with 
the movement of the carbon for the 8-mm "Suprex" combination than 
for the 1 1-mm angular- trim non-rotating carbon combination. It is 
also evident that with the movement of the positive carbon the dis- 
tribution of light changes much more rapidly for the "Suprex" carbons 
than for the 11-mm carbons in their respective lamp systems. For a 
decrease from the maximum of 10 per cent in the total light on the 
screen, the ll-mm carbon in its optical system can move over a dis- 
tance along its axis approximately 60 per cent greater than that re- 
quired to give the same reduction with the 8-mm "Suprex" carbon. 
For a range in the side-to-center distribution ratio of light on the pro- 
jection screen between 65 and 75 per cent, the 11-mm carbon with its 
optical system can move twice the distance of the 8-mm "Suprex" 
with its optical system. In other words, this 1 1-mm carbon with the 
optical system suggested gives a light on the projection screen which 
is about twice as stable with respect to carbon movement, etc., as the 
8-mm "Suprex" carbon in the commercial lamps of today. At the 
distributions indicated by the dotted portions of the curve in the fig- 
ure, the 8-mm "Suprex" gives non-uniform color over the screen, 
whereas the screen distribution with the 11-mm carbon is satisfac- 
tory over the entire range shown. 

This greater "flexibility" of the 11-mm carbon with its optical sys- 
tem is at least partially explained by reference to Fig. 2, where it is 
shown that the magnification of the optical system for the 11-mm 
carbon is so chosen as to use a portion of the crater which is relatively 
uniform in brilliancy. With a 5.2 X magnification of the crater of 
this carbon, the edge of the useful portion of the crater which is nor- 
mally imaged on the corners of the aperture shows a reduction of only 
15 per cent in brilliancy compared with the center; whereas with the 
8-mm "Suprex" crater and 6.6 X magnification, this reduction is 
25 per cent. 

With the increased amount of light from the 11-mm carbon and 
//2.0 optical system indicated in Table III, there will be an increased 
amount of heat on the film. Measurements have shown that the 

250 JOY, LOZIER, AND SIMON [j. s. M. P. E. 

spectral distribution of radiant energy with this 11-mm positive car- 
bon is similar to that of the other high-intensity arcs, and therefore 
the radiant energy falling on the film is approximately in proportion 
to the light passing through the film aperture. Heat measurements 
have been made with a blackened receiver fastened to a thermocouple 
and mounted in the plane of the film aperture. These indicate that 
this 11-mm carbon burned at 115 amperes in the//2.0 reflector lamp 
will put about 60 to 70 per cent more radiant energy on the film aper- 
ture than the conventional mirror or condenser type systems indi- 
cated in Table III. It may be advisable to make special provision to 
take care of this increase in heat at the film aperture. Methods for 
doing this are well known to the industry. 

A further idea of the possibilities for motion picture projection with 
this 11-mm carbon and the//2.0 reflector system is obtained from a 
consideration of the dimensions of the screen which can be ade- 
quately illuminated. Combining the data of Table II with the trans- 
mission of the film shutter, it is seen that about 8000 lumens can be 
put on the screen with the film shutter running and with no film in the 
gate. Under the above conditions the recommendations of the Pro- 
jection Practice Committee 12 call for a screen brightness of 7 to 14 
foot-lamberts. Assuming a diffusing screen of 0.75 reflectivity, it is 
possible to illuminate a screen approximately 34 feet in width to the 
minimum recommended brightness and 24 feet in width to the maxi- 
mum recommended brightness. This is about 60 to 100 per cent 
greater screen area than can be illuminated to this level by the reflec- 
tor or condenser lamp systems in common use today. 


1 BASSETT, P. R.: "The High Power Arc in Motion Pictures," Trans. Soc. 
Mot. Pict. Eng., No. 11 (1920). 

1 BRNFORD, F.: "The High Intensity Arc," Trans. Soc. Mot. Pict. Eng., 
No. 24 (1925), p. 71. 

8 BASSETT, P. R.: "The Progress of Arc Projection Efficiency," Trans. Soc. 
Mot. Pict. Eng., No. 18 (1924), p. 24. 

4 STARK, S.: "Reflector Arc Projection Some Limitations and Possibilities 
in Theory and Practice," Trans. Soc. Mot. Pict. Eng., No. 23 (1926), p. 94. 

5 JOY, D. B., AND DOWNES, A. C.: "Direct-Current High-Intensity Arcs 
with Non-Rotating Positive Carbons," /. Soc. Mot. Pict. Eng., XXII (Jan., 1934), 
p. 42. 

6 JOY, D. B., AND GEIB, E. R.: "The Non-Rotating High-Intensity D-C Arc 
for Projection," J. Soc. Mot. Pict. Eng., XXTY (Jan., 1935), p. 47. 

7 BASSETT, P. R.: "The Electrochemistry of the High-Intensity Arc," 
Trans. Amer. Electrochem. Soc., XLIV (1923), p. 153. 


8 HARDY, A. C.: "The Distribution of Light in Optical Systems," J. Frank 
Inst., 208 (Dec., 1929), p. 773. 

9 HARDY, A. C. : "The Optics of Motion Picture Projectors," J. Soc. Mot. 
Pict. Eng., XIV (March, 1930), p. 309. 

10 COOK, A. A.: "A Review of Projector and Screen Characteristics and Their 
Effects upon Screen Brightness," J. Soc. Mot. Pict. Eng., XXVI (May, 1936), p. 

11 We gratefully acknowledge the services of the Bausch & Lomb Optical Com- 
pany in supplying the //2.0 experimental reflector and 4-inch //2.0 experimental 
Cinephor type projection lens used in these tests. 

12 Report of the Projection Practice Committee, J. Soc. Mot. Pict. Eng., XXIX 
(July, 1937), p. 39. 



Summary. Up to a recent date, background process work had been limited to 
the size of a picture that could be successfully illuminated through a single projecting 

The origination of a combination of projectors superimposing identical prints of the 
same background on the screen simultaneously compounded the light delivery of a 
single machine and therefore greatly expanded the scope of background process photog- 
raphy for natural color and black-and-white. 

In discussing the Warner Brothers' compound or triple-head pro- 
jector, the author would like to quote from the Academy Award 
which this apparatus received for 1938: "for pioneering the develop- 
ment, and for the first practical application to motion pictures of the 
triple head background projector" and give a brief history of the 
circumstances that led to the inception of the principle of compound 
projection in process work, along with some of the steps by which this 
principle was brought to mechanical application to motion picture 

This discussion will be general, since the mechanical details have 
already been described comprehensively and excellently by A. F. 
Edouart. 1 

In the fall of 1937 Warner Bros. Pictures, Inc., scheduled a picture 
Gold Is Where You Find It which was to be photographed in Techni- 
color. The story was a melodrama of the early days of hydraulic 
mining in California, and had as its climax a sequence involving the 
flooding of a huge valley wherein were conducted these mining opera- 
tions. For obvious economic reasons a great many of the scenes of the 
climax had to be done by "trick work" miniatures, and process pro- 
jection shots utilizing the miniature backgrounds. 

* Presented at the 1939 Fall Meeting at New York, N. Y.; receired October 
12, 1039. 

** Warner Bros. Pictures, Inc., Burbank, Calif. 




In planning and designing our process shots it soon became 
apparent that to achieve the required scope and dramatic effects to 
put the sequence over, we were faced with an apparently insur- 
mountable barrier. The size of picture that had been successfully 
projected and rephotographed in color at that time was so small as 
to be practically useless in obtaining the scenes that the story de- 

FIG. 1. Rear view, showing lateral panning screw at bottom of base; also 
swiveled exhaust pipes overhead. 

The only development at that time was the single projector, 
equipped with the high-intensity and sundry types of light-sources 
which, in color process work, delivered as a maximum a picture 
9 X 12 feet in size. 

Our foreground settings consisted mainly of flumes, hydraulic guns, 
miners' shacks, etc., all too big to be identified properly with a picture 
so small. We were therefore forced to attempt to break the limita- 
tions which bound us. 

Reasoning that the practical maximum in light delivery had been 
reached through the single projector, the thought occurred that, 



tf. S. M. p. E. 

provided additional projectors could be superimposed over the one 
to compose a single image, the brilliancy of the picture should com- 
pound respectively and, by ratio, expand to the size required. After 
discussion relative to the amount of light needed we determined to 
attempt to superimpose three projectors because, in theory, this com- 
pounding of light would be enough. 

FIG. 2. Partial side view, showing vertical tilt device; 
also focusing handwheel in center. Rear view of mirror 
and adjustment screws. 

We recognized at once that the problem of superimposing projec- 
tors was mainly a problem of parallax. Being faced with the necessity 
of using our existing equipment, which by its very size, when placed 
side by side, would introduce an impossible parallax condition, we 
were forced to devise a "bread-board" set-up. 



This set-up was a T-shaped base of channel-iron, on the stem of 
which was mounted a center projector and on the two side arms 
of which were other projectors at right angles to the central machine, 
casting their pictures in the same plane through reflecting mediums 
consisting of 4-inch optical prisms. These prisms were separated 
laterally by a minimum distance that would clear the beam of the 
central machine, thereby making the three projecting beams originate 
from an area sufficiently small to cause no practical problems of 

FIG. 3. 

Quarter front view, showing means of rotating the projector and 
optical bedplate about the optical axes of side projectors. 

During the period of developing and applying this crude model 
to actual production conditions we were faced with many difficulties 
before it became practicable. Our first consideration, and the basis 
upon which the entire success of the experiment rested, was the ques- 
tion of whether our standard projector movements would prove steady 
enough for the absolute superimposition required. This, thanks to 
the precision of the standard Bell & Howell movements that we use, 
proved to be more than adequate. Our first projection of identical 



{]. S. M. P. E. 

charts on the three machines, while not superimposed absolutely as to 
position, showed no detectable creep or crawl. The foundation for 
practical application was there. 

The next problem was the development of a micrometer adjust- 
ment for positioning the reflecting mediums of the outboard projector 
beams both laterally and vertically so that their images could be 
superimposed upon the image of the central beam with a minimum of 
time and trouble and, once superimposed, remain fixed. 

FIG. 4. Front close-up view, showing lens mounts, mirror holders, and 
optical bench. The Bell & Howell camera movement can be seen through 
the window in the left-hand head; sync motor on right-hand head. 

The type of mounting was solved, but we found that the light loss 
and the weight of the 4-inch prisms that we first used hindered us in 
many ways. After experimentation, we ended with our present 
type of reflecting medium which is a fused quartz optical flat, sur- 
faced to l /J wavelength, aluminized first surface by the General Elec- 
trical Co. hardening process. 

With this "bread-board" set-up we successfully completed our 
assignment of Gold Is Where You Find It in the early part of 1938, and 
through actual production working conditions were able to formulate 
all the requirements that eventually went into our finished triple-head 



A major requisite for production conditions was that the entire 
unit could be adjusted easily laterally and vertically to position the 
picture on the screen without inaccuracy of superimposition. To 
meet this requirement a gun-mount type of base was designed which 
swings laterally around an axis equidistant from the three lenses and 
rotates vertically about the optical axis of the two outboard lenses, 
thus never altering the focal distance of any of the lenses from the 
screen during the necessary "positioning" of the picture for the shot. 

Another major requisite was rigidity of all the component parts in 

FIG. 5. Front view, showing division of lamp brackets from main base. 
The vertical load is carried on rollers in contact with the track forming the 
rim of the floor base. The projectors and optical bench are integral with the 
main rotating base. 

relation to the inertia of the base. In order to avoid any disturbance 
to the projector-lens base and consequently to the rigidity of the 
optical components, when the lamp houses were being manipulated 
all lamp-house mountings were separated from the portion of the 
base that contains the optical parts. The lamps, however, were 
attached to the sub-base so that they would stay in their optical 
position while the unit is being rotated. 

Arriving at the design of the optical base or optical bench, we 
found through developments of focusing apparatus that the whole 
set-up would be far more rigid if the lenses were solid to the bench 


and the projectors moved back and forth on generous dovetailed 
ways to obtain focus on the screen. We therefore have rigid lens 
mounts dowelled to accept matched sets of lenses so that they are 
always the same focal distance from the screen, thereby eliminating 
any adjustment for superimposition when changing the size of the 

These problems that we solved and incorporated into the finished 
product constitute the main features of the triple-head as it differs 
from the standard single unit. In the construction of the finished 
machine we were fortunate in having had a thorough working ex- 
perience with the "bread-board" set-up, and because of this we were 
able to design and build a finished product quickly without any 
further modification. 

In the practical application of the finished triple-head projector to 
working conditions of motion picture production, we bent every 
effort to avoid delays in the actual lining up and shooting of scenes 
during the time an expensive production company was on the set. 

It was our hope to be able to shoot a sequence of process shots 
with the same ease and expediency as was formerly done with the 
single unit. Mobility had been an important feature of the single 
unit. To gain this for the triple-head and at the same time not sac- 
rifice its rigidity we placed it on a concrete foundation in a sound-proof 
booth which operates somewhat like the turret of a battleship. This 
permits us to have several screens set up within the radius of the ro- 
tating field of the booth. 

Results during a period of constant use have proved that the 
triple-head is as practicable and economic as was the single unit. 
While the origination and completion of compound background 
projection came through a necessity for increasing the scope of process 
shots in color, it has proved through usage an important development 
in black-and-white work as well. Shots of a magnitude hitherto im- 
possible have been successfully completed. Effects where great 
depth of focus is required by stopping down on the photographic lens 
have been made. Whatever had been done before its creation was 
enlarged in proportion by the use of the triple-head projector. 


1 EDOUART, A. F.: "Paramount Triple-Head Transparency Process Pro- 
jector," J. Soc. Mot. Pict. Eng., XXXIII (Aug., 1939), p. 171. 


Summary. The various functions of motion picture theater auditorium lighting 
are discussed. Particular analysis is made of the lighting which is used during the 
period in which the motion picture is projected. Past and present lighting practices 
in this respect are explained. The advantages and disadvantages of these practices, 
and a new type of lighting are discussed. It is proposed that the illumination levels 
of the interior surface of the auditorium be at greater levels than have been heretofore 
found to exist. A definite relationship between the screen brightness and that of the 
auditorium surfaces is indicated as desirable. Recent tendencies toward higher 
screen brightnesses have made a very low intensity lighting in the auditorium much 
more undesirable, and therefore have made it more important to arrive at a new 
solution for motion picture auditorium lighting. The realism of the projected picture 
can be considerably heightened by proper surface illumination. Controlled reflected 
light coming from the screen and re-reflected from the interior surfaces is discussed 
as a medium for lighting. 

The auditorium of the motion picture theater is still being illumi- 
nated by the same methods, more or less, used in lighting the audito- 
rium of the stage theater. Yet, the lighting requirements in the two 
cases are decidedly different. These differences are best understood 
when the shapes and sizes of the auditorium and the motion picture 
in the one case, and of the auditorium and stage opening in the other 
case, are compared. Fig. 1 shows the amounts of interior surface vs. 
the motion picture and stage opening areas within the spectators' 
range of vision. Due to the necessarily oblong shape of the motion 
picture auditorium and the necessarily square shaped character of the 
stage theater auditorium, there is an important difference between the 
ratios of the interior surface areas and the areas comprising the motion 
picture or the stage opening. In each case there is a brightly illumi- 
nated area in the one, the motion picture; in the other, the stage 
opening. Note that the highly illuminated area in the stage theater 
occupies 3 /6 of a 60-degree horizontal range of vision, and in the case 

* Presented at the 1939 Fall Meeting at New York, N. Y. ; received Oct. 16, 

** New York, N. Y. 




(J. S. M. P. E. 

of the motion picture only l / 6 of a 60-degree horizontal range of vision. 
This comparison shows that in the stage theater the surface area with- 
in the spectators' range of vision but outside the highly illuminated 
area is relatively small and therefore could remain without illumina- 
tion or with optional illumination without any detriment to the proper 
enfolding of the stage performance. But in the case of the motion 
picture auditorium, the surfaces within the range of vision occupy 
a great proportion of the entire field, making it important to illumi- 
nate these surfaces so that they bear an appropriate relation and set- 
ting to the illuminated motion picture. 



Motion picture theater Stage theater 

FIG. 1. Interior surfaces within range of vision. 

For the same reasons that make it necessary to illuminate the 
auditoriums differently, it is also necessary to consider the design of 
the interior surfaces for each case. The decoration of the stage 
theater may be optional, but in the case of the motion picture theater, 
fixed traditional architectural decoration must be completely elimi- 
nated from the interior surfaces seen in combination with the motion 
picture. These surfaces must be neutral, and any divergence from 
the utmost simplicity in their treatment must be limited to changes 
in light intensity in the various parts for the purpose of creating a 
suitable setting for the picture. 


Past practices have totally ignored this approach to motion picture 
auditorium lighting, giving undue emphasis to irrelevant decorative 
lighting and thereby detracting from, rather than adding to, the 
effectiveness of the picture presentation. 

The lighting of the motion picture auditorium can be separated into 
three different and distinct functions : 

(1) Traffic Lighting. To permit patrons to move about safely to and from 
the seats. 

(2) Intermission or Decorative Lighting. Used solely for effect, during the 
waiting periods before and after picture showings. 

(3) Complementary Auditorium Surface Lighting. Used during the picture 
projection periods. 

This paper deals primarily with the third type mentioned, but a few 
remarks regarding the first two functions are important at this point. 

Traffic light should be controlled so that all the light intended for 
such purpose should fall upon the floors of the aisles and on the seated 
audience, avoiding any spill of light upon the wall and ceiling surfaces 
that come within the spectators' range of vision while viewing the 
picture. It is inadvisable to allow traffic lights to play upon such sur- 
faces because the intensity and placement requirements of traffic 
light and other lights used during picture projection periods are dis- 
tinctly different. Many successful methods of confining lighting to 
restricted areas have been developed and may be applied to traffic 

Intermission or decorative lighting may be optional, provided 
that such lighting is not used while the picture is being projected and 
that the lighting devices are concealed or rendered unobtrusive if 
within the range of vision while viewing the picture. The illumi- 
nated picture, if allowed to be, is the chief source of illumination of the 
motion picture auditorium during the projection period. Lack of 
recognition of this fact is the reason why the lighting problem has 
never been approached properly. Most screens have diffusive sur- 
faces and screen illumination levels have increased considerably, with 
the result that a considerable amount of light is reflected from the 
screen to the interior surfaces of the auditorium. Hence the problem 
of lighting is concerned with both the proper absorption or re-reflec- 
tion of this light as well as with the introduction of secondary light- 

Any contention that the use of light reflected from the screen should 
not be encouraged is not justified because it is exceedingly more ex- 

262 B. SCHLANGER [J. S. M. P. E. 

pensive to use instead completely secondary lighting sources for the 
illumination. Furthermore, secondary sources of illumination can 
not, within practical limits, be made to supply the desirable "syn- 
chronous" type of lighting to be described later in this paper. 

There are three or four possible theories of illumination during the 
picture projection period, leading to entirely different psychological 
effects upon the viewer. It may be contended that the entire audi- 
torium should approach an apparent total blackness, leaving the pic- 
ture in sharp contrast to its surroundings. In this case the necessary 
traffic lights, exit signs, and the screen light re-reflected from the 
audience is very likely to destroy the supposed illusion and in this 
instance, of course, the light reflected from the screen is supposed to 
be completely absorbed. To carry out the "dark" idea to its full ex- 
tent, a major part of the walls and ceiling would have to be of an al- 
most black, flat color, and a considerable number of the fixtures for 
traffic lights would have to be built in, to concentrate the light upon 
seats and the aisle floors. Yet, even if such a black effect were at- 
tainable economically, its desirability is questionable. A screen 
brightness of 7 to 14 foot-lamberts has been found desirable for the 
picture, and with such a screen brightness the dark auditorium sur- 
faces and screen surroundings contrast objectionably with the picture. 
This is so in the case of the black-and-white picture, and more so 
with color pictures. When daylight or other bright scenes are shown, 
the dark surroundings become even more undesirable. Therefore, 
the dark auditorium may be psychologically unsuitable as well as a 
factor in visual fatigue due to the strong contrast between the pic- 
ture and the surrounding darkness. 

Another theory would have the ulterior surfaces absorb most of 
the reflected screen light so that only the illumination from secondary 
sources would be apparent on the interior surfaces; but the cost of 
electric current and the installation expense makes this method un- 
economical when the lighting design is carried out properly. The 
method requires the use of indirect lighting, which is highly ineffi- 
cient. The lighting must of necessity be fixed in intensity and color, 
with the result that an unfavorable effect is produced when contrast- 
ing dark and light scenes are projected on the screen. However, this 
scheme does mitigate somewhat the visual fatigue which may be ex- 
perienced while viewing a bright picture in dark surroundings. 

Some experimental work has been done recently with the idea of 
causing the intensity and color of the secondary sources of light to 


change in conformance with the screen picture. This method will 
prove even more costly than the previous method, since elaborate 
electrical controls will be needed to synchronize the lighting of the 
screen and the interior. The scheme has all of the disadvantages of 
the music cue sheets that had to be supplied to each exhibitor in con- 
nection with the silent pictures; or of requiring each exhibitor to 
evolve a new light-synchronization scheme with each picture involved. 
These disadvantages indicate the impracticability of such an idea. 
The exhibitor should have no more to do than to project the picture 
properly and to let the picture create the mood. Certainly, the artist 
who is responsible for making the picture would not like to anticipate 
a thousand individual and different interpretations of illumination 
settings for the presentation of the picture. 
The successful lighting method must 

(1) Be reasonably inexpensive to install. 

(2) Cause the least visual fatigue while viewing a picture. 

(3) Create a desirable psychological effect. 

(4) Provide safe traffic lighting. 

While the first two requirements are of basic importance, the third 
is one that goes beyond the field of lighting and tends to deal more 
with the intended mood of the picture. Yet, since it is the lighting 
that creates the setting for the picture , all the requirements must be 
thought of as one and the lighting problem must embrace them all. 
The psychological problem nevertheless is the important one, because 
it determines the extent to which the picture will be enjoyed. 

The lighting methods previously discussed do not solve all the 
problems involved, especially that of creating the appropriate psy- 
chological mood, and a method is proposed by the author that would 
prove economical, relieve eye-strain, and make the picture presenta- 
tion more realistic. The patron would not feel picture-conscious, 
and would more easily come to feel that he existed in the actual scene 
depicted upon the screen. The effect is achieved by a lighting scheme 
that eliminates any contrasting framework around the picture. 

To achieve this effect, three steps must be taken: (1} All the in- 
terior surfaces, both ceiling and walls, seen in combination with the 
screen within a range of vision, must be in a white or nearly white 
color. (2) All such surfaces must be broken up into angularized 
planes or be of such surface textures that their brightness to the 
audience, under the light reflected from the screen, may be appropri- 


ately controlled. The surfaces are designed so that in certain parts 
they will appear brighter than in other parts. The apparent bright- 
ness at any point is controlled by re -reflecting the light from the screen 
toward either the cheeks or the eyes of the patrons in the desired pro- 
portions. Light reflected away from the eyes of the spectators is 
also helpful because it helps to increase the traffic lighting level. As a 
solution to the third step the use of synchronous screen border illumi- 
nation, contiguous to the screen, is suggested. The area immedi- 
ately around the picture is a transitional lighting area between the 
screen and interior auditorium surfaces the areas blend with one 
another, and as a whole are blended to the screen picture itself. The 
entire scheme of lighting is unified and related in all its parts. 

The result is a decided simplification. The lighting in the area 
adjacent to the picture is synchronous with the lighting that happens 
to occur at the edges of the picture. The surfaces beyond this ad- 
jacent area pick up their light in small amounts from the entire pic- 
ture; but as the distance from the screen increases the apparent 
brightness of the interior surfaces becomes lower, leaving them in 
darkness in areas outside the range of vision of the spectators. 

It is contended that present screen maskings and lighting and archi- 
tectural embellishments on these important surfaces in the interior 
are all distractions and, therefore, undesirable. A word of caution is 
appropriate here; viz., the mere setting up of a screen in a white 
auditorium not designed properly will be even more distracting than 
the dark auditorium, or an auditorium with secondary illumination, 
because the screen light will re-reflect on the white surfaces in an un- 
controlled and disturbing manner. 

With this method properly developed, the secondary light-sources 
needed for moving about the theater could be limited to small pilot- 
lights made to hug the floor in aisles, passageways, etc.; and even then, 
these lights would be obvious only when especially dark sequences 
appear on the screen. As to emergency lighting, it is extremely 
simple to arrange emergency lights that would automatically switch 



Summary. Special slide-changing projectors were designed and built for the 
Kodachrome exhibit in the Eastman building at the New York World's Fair. Eleven 
machines are synchronized so that panoramic scenes one hundred and eighty-seven 
feet long may be shown. Indexing of the slides is controlled by notches in a sound- 
film so that the entire program is automatic. 

The slides in each machine are arranged in two rows, and the machines each have 
two gates and two complete optical systems. All the slides in one row are rigidly 
bolted to a ring-gear forty-eight inches in diameter. For each new picture one of the 
two ring-gears is spring-indexed into a new position. An optical compensator 
geared to the ring-gear corrects for any inaccuracies in indexing, and the image is 
optically "dowelled" on the screen. 

The main feature of the Eastman Kodak Company's exhibit at 
the New York World's Fair is the Cavalcade of Color, a showing of 
projected Kodachrome slide pictures. 

The audience faces a semicircular faceted screen which is 187 feet 
long and 22 feet high. Above and behind the audience is a projection 
room which houses eleven special projectors. No operator is in this 
room. Day in and day out the only act performed by any operator 
is that of threading and running a notched sound-film. The first 
notch on this film dims the house lights; the last brings them up for 
an intermission. Other notches between these two put on the show. 
At times, synchronized with voice and music, the pictures projected 
on the screens change from individual shots of homely interest to 
complete panoramic views of nature's beauty spots. Altogether, 
more than two thousand individual screen pictures can thus be 
shown in a complete cycle. 

Possibly the most important single factor in making the show auto- 
matic is the use of incandescent lamps as light-sources. The individ- 
ual pictures are approximately one inch by one and a quarter, or 

* Presented at the 1939 Fall Meeting at New York, N. Y. ; received December 
27, 1939. 

** Eastman Kodak Co., Rochester, N. Y. 




LF. S. M. P. E. 

somewhat smaller than either the standard Retina or Bantam frame 
sizes. To obtain enough light for the 213X magnification on the 
screen, it was necessary to have a special lamp made and to design a 
high-efficiency condenser and objective system. The lamp is a 

FIG. 1 . General view of front of automatic slide projector, showing 
double projector system. 

2500-watt, biplane, coiled coil, differentially wound, pressure-filled 
tungsten lamp which operates at a high color-temperature (3450K). 
The condenser system is one quite free from spherical aberration, and 
consists of a piano-parabolic collection lens, two piano-spherical ele- 
ments, and a piano-spherical field lens, the combination of which 


images the filament to fill the diaphragm plane of the 4V2-inch//2.0 
objective. The actual amount of light obtained with this system is 
2250 lumens, giving better than six foot-candles illumination on the 
17 X 22-foot screens. 

Some of the individual pictures used in the show are projected for 
periods as long as 20 seconds. For this reason it was found necessary 
to cool the film in the gate. Between the lamp and the gate a 4- 
inch ferrous sulfate liquid cell is used. Sulfuric acid has been added 
to the ferrous sulfate solution to retard deposition of oxidized material. 
The gate itself is enclosed by glass, and filtered air cooled to 50F is 
blown over the film at a velocity of approximately 7000 feet per min- 
ute. The lamp is cooled by air at room temperature, circulated at 
the rate of approximately 100 cubic-feet per minute. The film, of 
course, is cemented on glass to prevent buckling and distortion in the 
hot gate. 

During the change-over from one picture to the next, it is not neces- 
sary for the screen to go black. A change can be made almost in- 
stantly from slide to slide, or one picture can be faded gradually into 
the next in a slow cross-dissolve since every one of the eleven machines 
projecting onto the eleven individual screens is double-barrelled (Fig. 
1). Each has two lamps, two gates, and two complete projection 
systems. During the time of projection from the left-hand gate of 
each of the eleven machines, new slides can be indexed into the right- 
hand gates; and as soon as they are positioned, by means of a shutter 
system, the left-hand pictures are faded out into the right-hand ones 
which allows new slides to be brought into the left-hand gates. The 
cross-dissolve shutters work in such a way that during the change-over 
the light to the screen remains constant. One picture can actually 
be substituted for another without alteration in the screen image if 
the registration of the two pictures is perfect on the screen. 

Many effect pictures can be shown if the registration is satisfactory. 
A color picture can be changed into a monochrome; it is possible to go 
from front lighting to silhouette lighting on the same subject; and to 
show complete panoramic scenes by day and by night. 

The big mechanical design job in making the projectors, then, was 
that of achieving precise registration. As noted, the film has to be 
cemented onto glass, so that it will not buckle or distort in the hot 
gate. The machines must be capable of handling some two thousand 
glass slides, keep them in order, and index them into position in such a 
way that the registration of the image on the screen seems to be perfect. 

268 F. TUTTLE [J. S. M. P. E. 

Many ways of doing this job were considered. The method finally 
selected has proved very satisfactory in actual practice. The film is 
cemented onto glass with the perforation holes located approximately 
with respect to the sides of the glass. The glass is then clamped and 
cemented onto metal carriers in an optical jig. The metal carrier has 
a machined face, which is used for locating it on the projector, and 
also gear-teeth which are used for registration purposes in the pro- 
jector. The result of these cementing operations is to put the metal 

FIG. 2. Front with lens-plate removed, showing gear-teeth on slide-car- 
riers and rotating plane-parallel plate for maintaining the picture stationary 
on the screen. 

gear-teeth on the film, accurately located with respect to the perfora- 
tion holes in the film. Once metal gear-teeth are on the film, they 
can be made to do some of the work of registration. 

Ninety-six of these slide carriers with their cemented slides are 
then bolted onto a ring-gear. The gear-teeth of the individual slide 
carriers form a continuous spur-gear 48 inches in diameter. The 
glass plates overhang the metal carrier and form a cylindrical drum 
of glass plates about 46 inches indiameter. The whole ring-gear and 
glass-plate drum assembly is mounted to rotate in a vertical plane. 

Mar., 1940] 



The lamp house and condenser system is inside the glass-plate drum, 
and the slides pass vertically through the gate. 

Vertical and circumferential faces of the ring-gear are accurately 
machined. The ring-gear assembly rolls through the gate on pre- 
cision ball- type roller bearings, riding on these vertical and circum- 
ferential surfaces. The bearings furnish accurate side-guiding for the 
pictures, which are rigidly attached to the ring-gear. They also force 
the pictures to stay in the focal plane. 

The whole ring-gear, metal film-carrier, and glass-plate assembly 

FIG. 3. 

Close-up of plane-parallel glass plate ; note the pinion-gear engaged 
by the ring-gear and the gear on the slide-carrier. 

weighs approximately two hundred pounds. In the indexing opera- 
tion this 200-pound drum assembly is rotated one ninety-sixth of a 
revolution and stopped in its new projection position. If good verti- 
cal registration by purely mechanical means is to be achieved the 
ring-gear must be stopped in exactly the right place. Conceivably 
this could be done (with a great deal of clatter) by dropping a me- 
chanical dowel-pin into an accurately located hole. Such a mecha- 
nism, however, would wear out rapidly. Instead of this mechanical 
d Dwelling of the entire ring-gear, it was decided to dowel the picture 



tf. S. M. P. E. 

optically in position. It is for this purpose that the gear-teeth were 
placed on the slide carriers (Fig. 2). 

The simplest form of "optical-dowelling" mechanism is the ro- 
tating plane-parallel plate. As was known from other uses of such a 
device an image of a moving object can be made to appear stationary. 
If this device could be used no matter where the ring-gear is stopped, 
or no matter whether it is still in motion during projection, the image 
can be made to appear in a fixed and stationary position ; and to use it 
all that has to be done is to engage a pinion driving the glass plate 
with the gear-teeth of the film carriers (Fig. 3). By using the glass 

FIG. 4. 

Close-up of spring element and overrunning clutch, which assures 
accurate, rapid indexing of the drums. 

plate one can afford to be YMO as inaccurate in stopping the ring- 
gear as without the plate. 

In the actual indexing operation of the drums, advantage has been 
taken of the fact that there are two drums and that the two are being 
indexed alternately. While one is moving the other is stationary. 
Let us consider these two drums as the two mass elements of a tor- 
sion-pendulum system. Connecting the two drums is a spring ele- 
ment wound up enough initially so that if one drum is held stationary 
when the other is released, it will swing to its stopped position (Fig. 
4). Mechanically it is caught and held there by an overrunning 


clutch which refuses to allow the drum to go backward. At this far 
position of its swing, it has stored up energy in the spring sufficient 
to index the other drum when it is released to its stop position, except 
for the frictional loss in the system. Actually, a small motor is used 
in the projector to keep the spring wound up, the power supplied by 
the motor being used to overcome the frictional losses of the system. 
This torsion -pendulum arrangement of the projector drums accom- 
plishes the indexing in a very short time, gives a purely sinusoidal 
start and stop, is very quiet, and accurate enough in its indexing throw 
to bring the new pictures into such a position that the glass plate can 
register the images. Because of the long projection periods used, the 
motor that rewinds the spring is not running continuously. It 
starts when the spring needs power and is automatically turned off 
when sufficient power has been supplied. 

The eleven machines as a group are electrically interlocked. 
Every time the left-hand drum on one machine indexes, the left-hand 
drums on all machines index. As soon as this indexing has taken 
place, the shutter mechanism starts a cross-dissolve which closes the 
right-hand lens and opens the left-hand lens. In general, this cross- 
dissolve occurs simultaneously on all projectors. However, the cross- 
dissolve can be made to occur in the machines in series. 



Summary. In the Bell Telephone Laboratories have been developed electrical 
circuits for the artificial production of speech. One form of the device is itself voice- 
controlled, thus differing fundamentaly from the Voder of the World's Fair which is 
controlled by keys and pedals. It has been christened the " Vocoder" or "voice coder." 

Many startling effects are possible when the code is varied, for the Vocoder then re- 
creates sounds quite different from those used by the person speaking. Cadences 
may become monotones, rising inflections may be turned to falling inflections, a vigor- 
ous voice may become a quaver, or a single voice may accompany itself at any desired 
musical interval thus converting a solo into a duet, etc. Also non-speech sounds 
may be coded into intelligible speech and instrumental music into vocal music. 

At the World's Fairs in New York and San Francisco great interest 
has been shown in the speech synthesizer of the Bell System exhibits. 
Known as the Voder, l or Foice Operation Z)monstrato.K, this device 
creates spoken sounds and combines them into connected speech. 
Its raw materials are two complex tones, a hiss and a buzz; selection 
of one or the other and its intensity and tone quality are controlled by 
an operator through a keyboard. 

The Voder is an off -shoot of a more extensive system, first demon- 
strated in an experimental stage some three years ago. That system 
analyzed spoken sounds, and then used the information to control 
the synthesizing circuit. At the time, the World's Fair displays were 
under consideration, so it was naturally perceived that the synthe- 
sizer, manually controlled, could be made into a dramatic demonstra- 
tion. Development was for a while concentrated in that field ; as a 
successful Voder became assured, attention was shifted back to the 
broader parent system. Shortly thereafter the system was named 
Vocoder because it operates on the principle of deriving voice codes 
to re-create the speech which it analyzes. The Vocoder is pictured in 
Fig. 1. 

* Presented October 17, 1939, before a joint meeting of the Society of Motion 
Picture Engineers with the New York Electrical Society, at New York, N. Y. 

** Bell Telephone Laboratories, New York, N. Y. 



Fig. 2 shows the overall circuit for remaking speech ; the analyzer is 
at the left and the synthesizer at the right. Electrical speech waves 
from a microphone are analyzed for pitch by the top channel and for 
spectrum by a group of channels at the bottom. 

In the pitch analysis the fundamental frequency, which for sim- 
plicity will be called the pitch, is measured by a circuit containing a 

FIG. 1. The Vocoder. . 

frequency-discriminating network for obtaining this frequency in 
reasonably pure form; a frequency meter for counting, by more or 
less uniform pulses, the current reversals therein; and a filter for 
eliminating the actual speech frequencies but retaining a slowly chang- 
ing current that is a direct measure of the pitch. (Unvoiced sounds, 
whether in whispering or the unvoiced sounds of normal speech, have 
insufficient power to operate the frequency meter.) The output cur- 



[J. S. M. P. E. 

rent of the pitch channel is then a pitch-defining signal with its current 
approximately proportional to the pitch of the voiced sound and 
equal to zero for the unvoiced sounds. 

There are ten spectrum-analyzing channels,* the first handling 
the frequency range 0-250 cycles and the other nine, the bands, 300 
cycles wide, extending from 250 cycles to 2950 cycles, a top frequency 
which is representative of commercial telephone circuits. Each 
spectrum-analyzing channel contains the proper band filter followed 
by a rectifier for measuring the power therein and a 25-cycle low-pass 
filter for retaining the current indicative of this power but eliminating 
any of the original speech frequencies. 

The operation of the analyzer is illustrated in Fig. 3 by a group of 











_n is s 



" 1 ~~ 


- , FILTER , 







FIG. 2. Schematic arrangement of the Vocoder. 

oscillograms taken in analyzing the sentence She saw Mary. To in- 
sure that the same speech was analyzed in obtaining the various os- 
cillograms, the sentence was recorded on a high-quality magnetic-tape 
recorder with reproductions therefrom supplying current to the 
analyzer. The speech wave input to the analyzer is shown in the line 
next to the bottom, while the output is shown in the other oscillogram 
traces ; the pitch-defining signal is at the bottom in the figure and the 
ten spectrum-defining signals in numerical order at the top. For con- 
venient reference the oscillograms are lined up together, whereas in 

* A 30-channel Vocoder covering the wide range of speech frequencies required 
for high quality has also been built and is being used as a tool in laboratory investi- 

Mar., 1940] 



the actual circuit the speech-defining signals lag about 17 milliseconds 
behind the speech input wave. The inaudible speech-defining out- 
put signals contain all the essential speech information as to the input 
wave, but it is to be noted that they are slow-changing and in this 
way correspond to lip or tongue motions, as contrasted with the 
higher audible vibration rates of the rapid-changing speech wave 
itself. The dropping of the pitch to zero for the unvoiced sounds sh 
and 5 is also readily seen. 

Fig. 3 gives an idea also as to the synthesizing process. In the 
analyzer the speech wave is the input and the eleven speech-defining 

FIG. 3. 

The original speech wave and an analysis of its components, ex- 
pressed by the variation of several direct currents. 

signals are the output; in the synthesizer the eleven speech-defining 
signals are the input and the speech wave the output. 

The steps in speech synthesis are indicated at the right of Fig. 2. 
The relaxation oscillator is the source of the buzz; and the random 
noise circuit the source of the hiss. The hiss is connected in circuit 
for unvoiced sounds and for quiet intervals. (In the latter case no 
sound output from the synthesizer results because there are no cur- 
rents in the spectrum channels.) When a voiced sound is analyzed a 
pitch current other than zero is received with the result that the buzz 
is set for the correct pitch by the "pitch control" on the relaxation 
oscillator; and at the same time the relay marked "energy source 

276 H. DUDLEY [j. s. M. P. E. 

switch" operates, switching from the hiss source to the buzz source. 

The outputs from the spectrum-analyzing channels are fed to the 
proper synthesizing spectrum controls with the band filters lined up to 
correspond. The power derived from the energy sources of the 
synthesizer in these various bands is then passed through modulators 
under the control of the spectrum-defining currents. The result is 
that the power output from the synthesizer is sensibly proportional in 
each filtered band to that measured by the analyzer in the original 
speech. From the loud speaker comes, then, speech approximately 
the same in pitch and in spectrum as the original. This synthetic 
speech lags the original speech by about 17 milliseconds due to the 
inherent delay in electrical circuits of the types used. 

In the present models of the Vocoder, control switches have been 
introduced which permit modifications in the operation of the synthe- 
sizer. Through the manipulation of these controls interesting effects 
are produced. Some of the possibilities were demonstrated by the 
author and his associate, C. W. Vadersen, before the Society of Motion 
Picture Engineers and other groups. In this presentation Mr. Vader- 
sen supplied by his own voice the incoming speech which was picked 
up by a microphone as shown in Fig. 2, while at the same time he 
manipulated the controls to produce desired effects. A remote-con- 
trol switch was also provided through which, for purposes of com- 
parison, the author could switch the microphone directly to the loud 
speaker and so let the audience hear how the speech would sound if it 
had not been modified by passing through the Vocoder. 

In these demonstrations comparison is first made between direct 
speech and the best re-creation that the Vocoder can make. Then 
by manipulation of dials and switches, speech is modified in various 
ways. Normal speech becomes a throaty whisper when the hiss is 
substituted for the buzz. Although the hiss is relatively faint, it is 
shown to be essential for discrimination as between church and 

Ordinarily the re-created pitch moves up and down with that of the 
original. If variation is prevented, the re-created speech is a mono- 
tone, like a chant. When the relative variation is cut in half, the 
voice seems flat and dragging; when the swings are twice normal, the 
voice seems more brilliant; when four times normal it sounds febrile, 
unnatural. The controls can be reversed so that high becomes low: 
the tune of a song is then unrecognizable, and speech has some of the 
lilting characteristics of Scandinavian tongues. Another control 

Mar., 1940] THE VOCODER 277 

fixes the basic value of the re-created pitch; if this is "fluttered" by 
hand, the voice becomes that of an old person. By appropriate 
setting of the basic pitch, the voice may be anything from a low bass 
to a high soprano, and several amusing tricks can be performed. In 
one of these, the basic pitch is set to maintain a constant ratio of 5 to 
4 to the original. This is a "major third" higher and harmonizes 
with the original. In two-part harmony, the demonstrator then 
sings a duet with himself. Connecting a spare synthesizer set for a 
3 to 4 ratio, he then sings one part in a trio, the others being taken by 
his electrical doubles. Finally with the basic pitch-control, he be- 
comes a father reprimanding his daughter; then the girl herself ; and 
then the grandfather interceding for the youngster. 

For the vocal-cord tones of the original, the Vocoder substitutes 
the output of a relaxation oscillator. But any sound rich in harmon- 
ics can be used : an automobile horn, an airplane roar, an organ. In 
some demonstrations, the sound, taken from a phonograph record, 
replaces the buzz and hiss inputs. Keeping careful time with the 
puffs of a locomotive, the demonstrator can make the locomotive 
puff intelligibly "We're start ing slow ly faster, faster, fas- 
ter" as the puffs come closer together. Or a church bell might be 
made to say "Stop Stop stop don't do that." A particu- 
larly striking effect is that of singing with an organ to supply the 
tones; while the words may be spoken, the demonstrator usually 
sings them to hold the rhythm, but it makes no difference whether his 
voice is melodious or not; the tonal quality comes only from the 
musical source. 

These demonstrations of taking a speaker's voice to pieces, measur- 
ing the pieces and then building a related new voice shows that the 
Vocoder has possibilities as a tool in the investigation of speech, since 
by its numerous controls important variables in speech can be iso- 
lated for study. As to the engineering possibilities which may grow 
out of the application of the principles employed in this device, it is 
hard to predict at the present time. The speech-defining currents, 
however, do have features of simplicity and inaudibility which may 
open the way to new types of privacy systems or to a reduction in the 
frequency range required for the transmission of intelligible tele- 
phonic speech. 

It is obvious that this remaking of the voice is of interest in sound 
recording. Voice can be added to any natural or artificially pro- 
duced sound. Thus the ringing of a bell may be made to convey 

278 H. DUDLEY 

intelligible speech without, however, losing the distinguishing fea- 
tures of the notes of the bell. New voices can be made extending 
beyond the limitations of human voices. On the one hand uniform 
changes can be projected into a number of voices to give a family 
characteristic while on the other hand the same voice can be given an 
almost unlimited variety by merely adjusting control dials. The 
voice changes are produced consistently and automatically without 
effort on the part of the talker. This apparatus gives promise of 
ultimately bringing to the sound in the talking motion picture a de- 
gree of flexibility comparable to that already available for the picture 
itself to enhance its enjoyment. 


1 DUDLEY, H., RIESZ, R. R., AND WATKINS, S. S. A.: "A Synthetic Speaker," 
J. Franklin Inst., CCXXVII (June, 1939), p. 739. 

2 DUDLEY, H. : "The Automatic Synthesis of Speech," Proc. Nat. Acad. Sci., 
XX (July, 1939), p. 377. 

3 DUDLEY, H.: "Remaking Speech," J. Acoust. Soc. Amer., XI (Oct., 1939), 
p. 169. 



Summary. A definition of a graininess coefficient G is given by the distribution 

g-te/G'^dx O f relative transparency fluctuations (&T/T m 

x) in emulsions. In practical units G = 1000G'. 

An instrument is described which measures the graininess of an emulsion sample 
directly in G. It consists of the graininess photometer which produces a microphoto- 
metric record in terms of x and an integrator which analyzes this record photoelec- 
trically in terms of the probability function and indicates G. There are no standards 
involved in the instruments except for the geometrically defined area which scans the 
sample and which is simply related to G. 

The G values obtained for different emulsions are described for constant density as 
well as the variation of G with the density. The transformation of G into G,, the 
graininess effective in sound reproduction, is given as G a = G-10~ D ; furthermore, 
the corrections necessary for G to obtain visual graininess impressions, are discussed. 
The latter varies with the illumination on account of deviations from Fechner's law 
for certain ranges of brightness. Their visual effects are predictable from G. 

The effect of scattering of the incident light within the emulsion upon the measured 
graininess is discussed. 

The observed graininess density functions are compared with those of other authors 
and agreement and deviations are found where they are to be expected. 

The relation between graininess and grain size is discussed with the conclusion 
that a priori no relation exists between the two qualities. 

The grain distribution is found to follow, in many cases, very closely the probability 

In a previous paper 1 the theoretical background for an objective 
and absolute determination of the graininess of photographic emul- 
sions has been discussed; moreover, the fundamental design of an 
instrument performing graininess measurements has been given. 
Meanwhile a graininess meter for practical use was developed which 
has been in continuous operation for more than a year in a large in- 
dustrial laboratory. Its construction and some of the results obtained 
with it shall be described henceforth. 

* Received February 13, 1940. 
** California Institute of Technology, Pasadena, Calif, 


280 GOETZ, GOULD, AND DEMBER []. s. M. P. E. 

Fundamental Principle,* The objective measure of the graini- 
ness, as previously proposed, makes the following assumptions : 

(a) The graininess is defined by the size and the frequency of 
occurrence of transparency fluctuations. This definition does not 
consider explicitly the average grain size but solely the grain distri- 
bution which for the same optical density involves the former im- 

(b) This frequency of occurrence of fluctuations can be adequately 
described in terms of the probability law (Gaussian function). The 
assumption of a distribution law for the occurrence of fluctuations 
is necessary, as without it the definition of a graininess coefficient is 
impossible which is at once descriptive of the frequency as well as of 
the size of transparency fluctuations. 

(c) An adequate approximation to the subjective impression of 
graininess is given by expressing the size of the fluctuations in terms 
of relative transparency fluctuations. If T m is the mean transparency 
of the emulsion and AT the absolute size of a particular fluctuation, 
AT/T m represents the relative size of the fluctuation which is dis- 
scriptive of the subjectively realized density variation (AD). The- 
approximation becomes faulty for large densities, however, graininess 
measurements are in general not interesting in such regions. 

The above assumptions result in the following definition of the 
graininess coefficient G' in terms of the Gaussian function for the 
probability : 


= f x 

T Jo 


where G' alone describes the degree of irregularity in the grain ar- 
rangement: Large values of G' thus indicate a large frequency of 
occurrence (probability) for large fluctuations, whereas for small 
values of G' the number of large fluctuations is small and that of small 
fluctuations is increased correspondingly. The units of G', as defined 
by the equation, prove to be too large for a convenient description of 
the graininess actually occurring in photographic emulsions. There- 
fore G = G' 10~ 3 is used as the practical unit. 

The establishment of this coefficient requires an instrument which 
measures and indicates the graininess in terms of G. In principle a 
microphotometric record of the relative transparency flustuations of 

* For theoretical details see ref. 1. 



an emulsion is taken, then this record is analyzed in terms of the 
probability law and the graininess coefficient is evaluated. The former 
operation is performed by the graininess photometer, the latter by 
the graininess integrator. 

The Graininess Photometer. The purpose of this instrument is to 
obtain a record of the relative transparency fluctuation (AT/T m ) of a 
sample of a photographic emulsion. The record is produced photo- 
graphically on 35-mm positive film. Each record has the length of 
one meter, which in turn can contain up to 10 5 fluctuations which 

5.' 41 2 A 

FIG. 1. Scheme of the graininess photometer: (The numbers refer to the 
various parts of each unit designated with a capital letter). 

number represents the upper limit of the combined resolving powers 
of the taking as well as the recording systems. 

A schematic drawing of the arrangement of the instrument is given 
in Fig. 1. The light-source for the microphotometric unit consists 
of a d-c tungsten arc lamp which is mounted adjustably in a ven- 
tilated housing. The light passes an achromatic condenser lens L4 
and a heat-absorbing filter L5, before entering the microscope M. 
Here the beam passes an iris diaphragm Ml upon which the image of 
the anode of L2 is focused. The iris Ml is adjustable by means of a 
gear transmission which is led to the front panel of the instrument to 
K3. The purpose of this iris is to adjust the illumination of the 



[J. S. M. P. E. 

emulsion without changing the current in the arc lamp (which would 
result in a change of the spectral qualities of the light). 

Behind the iris a 90-degree prism M2 is mounted which deflects the 
light-beam into the optical axis of the microscope. The light then 
passes the aplanatic condenser M3 (0.6 numerical aperture), whence 
it passes through the emulsion sample which is mounted upon the 
photometer stage M4 into one of the two objectives M5 or M6 which 
are mounted upon a revolving nose piece. M5 is used for the survey 
of the sample. It is achromatic 5X whereas M6 is used lor the taking 
of the actual record and is apochromatic 20 X (0.65 N.A.). 

R 7 R 6 R 

M 13 


M 1 


FIG. 2. 

Photograph of right side of graininess photometer half assembled. 
(The designations refer to those in Fig. 1.) 

The stage which facilitates the promotion of the mounted sample 
during recording is supported upon the base of the microscope at 
three points, two of which lie in the groove M7 (Fig. 2), the third upon 
a pivoted steel disk M8. The former supports are given by polished 
steel balls which roll with an interposed spacer in the groove M7 and 
carry the stage S in a groove. The disk M8 supports the stage at a 
planed surface. 

The motion of the stage 5 on the base of the microscope is controlled 
in the following way: A micrometer spindle rests in two adjustable 



bearings in the base exactly parallel to M7. This spindle is rotated 
(at 3 rpm) by a motor drive. The thread of the spindle (10 threads 
per cm) engages into a worm gear which rests on pivots attached to 
the stage. The worm gear can be clamped so that its free motion 
relative to the stage is arrested; consequently the stage is propelled 
against the base if the micrometer spindle is turned. If the clamping 
mechanism, however, releases the worm gear, the stage does not move 
as only the worm gear is turned by the spindle. The clamping mecha- 

R 6 

R 7 

R 3 

R 14fH 

C 11 

FIG. 3. Photograph of left side of graininess photometer half assembled. 
(The designations refer to those in Fig. 1.) 

nism is operated by a flexible wire control from a push button (Z9, 
Fig. 4) on the switchboard ; at the same time it controls, through two 
mercury switches, the motor and the recording light (see later). 

The stage is driven by a resiliency mounted induction-motor which 
is coupled flexibly with a transmission assembly of spiral gears 
mounted in a box TR, coupled flexibly with the spindle M9 (Figs. 1 
and 2). 

The propelling mechanism of the stage moves the sample at a 
uniform speed of 3 mm per minute during recording; it is adjusted to 

284 GOETZ, GOULD, AND DEMBER [j. s. M. P. E. 

move the stage over exactly 1 cm which corresponds to a length of a 
record of slightly more than 1 meter; when the stage has travelled 
over this distance, the clamping device S5 is released and the stage 
is pushed back, by means of a spring, into its starting position. The 
impact of the backward motion on the microscope is damped by a 
shock-absorbing device. In this way it is possible to start the motion 
of the stage at any position of the latter, by the engagement of the 
clamping mechanism, and to release it automatically, and return to 
the starting position, after the stage has travelled through a distance 
sufficient for one record. 

On the upper side the stage carries an adjustable cross stage which 
clamps the emulsion sample and is arranged for holding of 1 X 3-inch 
plates. Its purpose is to facilitate the % selection of a perfect region of 
the emulsion sample which is free from dust particles, scratches, etc. 
The sample is embedded in Nujol and mounted between thin plane 
glass plates. 

Above the objectives M5 or M 6 the light passes a semitransparent 
mirror M10 (Fig. 1) which splits it into two beams. The reflected 
beam enters the barrier-layer photocell Mil (Lange-type) whereas a 
fraction of the transmitted beam passes through the ocular M12 
( 1 5 X ) . The latter beam then enters a system of mirrors : The small 
mirror M13 (Fig. 1) reflects the light horizontally upon the mirror 
M14 from where the major part is reflected upon M15 and from there 
upon a ground glass screen M 19 mounted into the front panel of the 
instrument (Fig. 4). All mirrors are first-surface mirrors (aluminum) 
on plate glass. 

The photocell Mil receives the light from the full optical field of 
the objective which is large enough to be independent of the local 
transparency fluctuations (graininess). The current produced by 
Mil is thus the measure for the mean transparency T m . This cur- 
rent is brought by a shielded cable to the mirror galvanometer (d'Ar- 
sonval-type) G of compact construction. Its light mark is focused 
upon and visible through the ground-glass screen G5 in the front 
panel (Fig. 4) where accordingly an image of the pointer on G3 ap- 
pears. A travelling mark can be adjusted with G6 for fixing the 
particular galvanometer deflection desired. 

A fraction of the light reflected from M13 passes through a hole 
drilled 45 degrees in the center of M14. Before it is received 
by the photoelectric cell A3 (vacuum type) it is restricted by the 
diaphragm M16, the aperture of which is proportional to the "stand- 


ard scanning area" a, or "standard integer" (see below). From there 
it passes through a short-focus lens M17 which spreads a diffused 
magnified image of the section of the emulsion transmitted through 
M16 upon the full cathode area of A3. 

The purpose of this arrangement is to measure relative transparency 
fluctuations of a small area of the emulsion (defined by diaphragm 
aperture/magnification) with a photoelectric cell and to be able to 
check the neighborhood of the recorded area of the emulsion all the 
time by visual observation of M19 (where the hole in M14 appears as a 
black circle in the center of the projected image). It is thus possible 

G5_ Z8 X5 G.6 XSJ Z10 

Z 6 K 3 Z 9 

FIG. 4. Photograph of graininess photometer fully assembled. 

to see whether or not accidental transparency fluctuations due to dust 
particles, etc., lie in the path of the record.* 

The power required for the recording of the relative transparency 
fluctuations is supplied by an amplifier A (Fig. 1) which is controlled 

* The photometer is arranged also for use as photomicrographic apparatus for 
emulsions. For this use M19 can be slid out of its position and replaced by a 
9 X 12-cm. photographic plate holder. As it is undesirable to have an image of 
the hole in M14, provision is made to slide M14 into a position where the hole 
becomes invisible. 

286 GOETZ, GOULD, AND DEMBER (j. s. M. p. E. 

by the fluctuations of the photocell A3. In order to avoid the com- 
plications involved in a d-c amplifier the light-beam after M17 is 
chopped at its focal point by a chopping disk A 2 which is driven by a 
small electromotor Al, mounted in a resilient cradle. The holes in 
the disk are arranged so as to produce an alternating light-beam of 
860 cps, the wave-form of which is nearly square since the focal point 
of M17 lies in the center of the holes. 

The amplifier is of the resistance-coupled audio-frequency type and 
is mounted in a carefully shielded separate unit which rests upon a 
welded steel frame above the lamp L and the driving motor MR. 
The output goes into a power stage and then through a transformer 
into a full-wave rectifier, where the current is rectified so that a direct 
current of 1720 fluctuations per second results. The gain of the 
amplifier is controlled by a variable potentiometer resistance which is 
adjusted through a flexible lead from the switchboard at K2 (Fig. 4). 

This current is led into the recording galvanometer R3 (Figs. 1 and 
3), which is of the liquid-damped oscillograph type and responds to 
vibrations up to 1800 per second. 

The optical path of the recording system is designed as follows: 
From the light-source Rl the beam passes the spherical condenser 
R2 which carries a slit; after reflection at the galvanometer mirror it 
passes through the diaphragm R5 and the system R6; it is then re- 
flected by a mirror and enters the camera vertically through a slit 
which can be closed by the shutter R12. The cylindrical objective 
Cll focuses an image of R2 upon the recording film. The width of 
this image is 10 microns, its length up to 20 mm. The fluctuating 
deflection of R4 thus causes a variation of the length of the slit image 
on the recording film. The zero position of the image on the center 
of the film is adjustable. The record is taken on 35-mm positive 
cine film. The capacity of the camera is 100 feet. 

The camera is loaded through the lid C (Figs. 1 and 3) on the left 
side of the instrument where the full reel C9 supplies fresh film. This 
is led over the friction drive CIS to the reel CIO upon which the ex- 
posed film is stored. For the purpose of adjustment of the image of 
the recording light-beam, the shutter R12 is provided, which prevents 
the exposure of the film. 

CIS, together with C9 and CIO, are driven by the motor MR, 
through the transmission TR through a flexible coupling which guar- 
antees a synchronous motion of the microscope stage and the record- 
ing camera, at a rate of 1 : 100. 


The electric power supplies are the following: 110-v a-c, 60-cycle 
for the motors and lights; 110-v d-c for the tungsten arc supplied by 
storage batteries ; 270-v from dry-cells for the amplifier and photocell 
and 6-v from a storage-battery for the filaments. Figs. 5 and 6 are 
reproductions of graininess records produced by the photometer 
they will be discussed later. 

The Graininess Integrator. The purpose of the graininess integrator 

FIG. 5. Graininess records of six different emulsions of nearly equal den- 
sity. On the right, photomicrographs of emulsions (300 diam.) for com- 
parison of objective and subjective graininess. 

is to analyze the graininess record of the photometer in order to de- 
termine the distribution of the relative transparency fluctuations 
quantitatively. This analysis requires of the integrator the measure- 
ment of the amplitude of each single fluctuation, furthermore the 
establishing of the frequency of occurrence of fluctuations of the same 
amplitude, i. e., the "counting" and "assorting" of up to 100,000 
A7"'s. This statistical calculation, if performed by the customary 
methods, would require the labor of several days or even weeks, 



[J. S. M. P. E. 

whereas the integrator performs these operations photoelectrically in 
a few minutes with the accuracy of an extended calculation. 

In principle the working of the apparatus is illustrated in Fig. 7 : 
The drum 1 rotates about its horizontal axis 12 and carries at its 
outer circumference a groove into which the record R is mounted. 

FIG. 6. 

Graininess records of a panchromatic emulsion for different 

The drum is driven by the motor M T through a friction drive. A 
small section of the record is illuminated by the projector 2, consisting 
of a light-source 21, a condenser 22, and a slit 23. The direction of 
the slit is parallel with the direction of the rotation of the record. An 
image of this slit is projected by the microscopic objective 24 upon 



the record. The light transmitted by R is received by the photocell 
25 built into the same frame with the projector. This frame 2 is 
mounted on a bed along which it can be moved normal to the direc- 
tion of rotation of the record. This motion is accomplished by the 
attachment of 2, to the micrometer spindle 3, which is, by a number 
of gear transmissions (not shown in Fig. 7) connected with the dials 
33 and 34. 


FIG. 7. Scheme of the graininess indicator. (The 
lower section shows a front view, on a larger scale, of the 
dials with three G-spirals on each.) 

The operation of this arrangement causes the taking of an average 
of the recorded fluctuations for each particular position of the slit; 
in other words, the light transmitted to the photocell through the rec- 
ord corresponds to an average transparency when the drum is ro- 
tated, equivalent to the light transmitted through a slit one meter 
long were the record at rest. This, of course, is true only if the speed 

290 GOETZ, GOULD, AND DEMBER (j. s. M. P. E. 

of the drum is high, relative to the period of the indicating galva- 

The photocell 25 is connected with a mirror galvanometer G whose 
light-beam is focused upon the plane of the dials 33 and 34 which face 
the front of the instrument. The antiparallel motion of the dials, 
facilitated by the gear transmission 32, is coupled with the motion of 
the spindle 3, so that each position of the dials corresponds to a cer- 
tain position of the slit on the record. The image of the cross-hair of 
the galvanometer moves thus gradually from the center of one dial 
to the center of the other dial when the slit moves from the black to 
the transparent side of the record, i. e., as the photo-current goes 
from zero to a value corresponding to the transparency of the clear 
part of the record. 

The dials are of semitransparent material upon which a family of 
spirals with numerals is drawn (Figs. 7 and 8). These spirals repre- 
sent the function of the probability law drawn in polar coordinates, 
where the angle represents the relative transparency fluctuation, and 
the radius vector the frequency of occurrence. Each spiral being 
drawn for a different G- value represents thus the geometrical locality 
("locus") of occurrence of the frequencies for a given value of G 
(graininess constant) . 

The analysis mentioned above is thus performed by moving the 
slit across the record and observing the resulting motion of the cross- 
hair on the dials: if the probability law is fulfilled, the cross-hair 
will be seen to "select" one of the spirals and follow it through both 
disks from one end to the other. The number attached to this par- 
ticular spiral (i. e., the G- value for which it was constructed) is the 
graininess coefficient of the emulsion from which the record was drawn. 

The dials are mounted behind the front panel which carries a 
shielded slit through which the part of the former is visible which is 
illuminated by the light-beam of the galvanometer. 

Although the operation of the instruments in principle appears to 
be rather involved, the taking and analyzing of graininess records is a 
matter of a few minutes which is a very short time compared to the 
labor necessary for a mathematical analysis of a record. It may be 
mentioned that this instrument can be applied also to the analysis of 
many other statistical problems. 

The Standards of the Graininess Meter. One of the most important 
problems in the definition of the graininess coefficient G and the de- 
velopment of the graininess n:cter is the question of the dependence of 


the results on standards of the individual instrument which are not 
directly a part of the definition of G. 

Aside from the undesirability of such instrument standards on ac- 
count of their variation with time and wear of the meter, the un- 
avoidable discrepancy of measurements taken with different instru- 
ments makes it highly desirable to reduce the number of individual 
instrument constants to a minimum. The previous description 
shows that they have been eliminated entirely with the sole excep- 
tion of the size (a) of the "standard scanning area" (or "standard 
integer"*) as defined by the area of the diaphragm M16 in Fig. 1, 
which determines the size at which a given grain fluctuation will ap- 

Photograph of front of integrator, showing the 
dials. (Front panel removed.) 

pear on the record. The larger a, the smaller the resolving power of 
the recorder as long as the section of the sample which is determined 
by a is large compared with the size of a single grain. 

The choice of a is practically limited toward lower sizes by the 
photoelectric sensitivities required which would render the instru- 
ment too delicate for practical use, and toward larger sizes by the 
lack of resolving power for graininess details. The area of M16 
chosen in view of these considerations corresponds to a radius of 15 
microns or an area of 0.0007 mm 2 on the emulsion. 

* This "standard scanning area" should not be confused with the area of the 
emulsion actually scanned during the taking of a graininess record, 

292 GOETZ, GOULD, AND DEMBER [j. s. M. p. E. 

It must be noted that, although the G-values obtained depend on 
the a, they can be compared easily with values G\ obtained with a 
different scanning area a\ through the relation G = Gi\/ai/a. 

The indications of the photometer, which depend on the sensi- 
tivity of photocells and amplifier, do not require standards as only an 
initial adjustment before operation is necessary, which consists in 
adjusting the gain of the amplifier and with it the resulting sensitivity 
of the recording system to a standard deflection of the galvanometer 
G2 (T m = 1) without interposition of a sample on the stage. This 
adjustment takes care of the relative sensitivity of all different ele- 
ments of the photometer. 

For similar reasons no standards are involved in the integrator, as 
the galvanometer deflections are always adjusted to the scale of the 
dials. The width of the integrating slit (23, Fig. 7) does not affect 
the evaluation of G except in that the width determines the lower limit 
of G accessible to measurement which is also the margin of error of all 
readings taken with it. As such a limit is already given by the size of 
a, the sensitivity of the recorder, and the imperfections of the profile 
on the record, the increase of the "resolving power" of the integrator 
does not need to exceed these limitations. 

The Graininess of Different Emulsion Types. Some of the results 
obtained with the graininess meter are described in the following: 
It has been shown previously that the graininess coefficient G was 
chosen in such a manner that its values can be expected to be repre- 
sentative of the subjective impression of inhomogeneity realized by 
the observer of a sufficiently enlarged section of a photographic emul- 
sion. The subjective impression in each special case, of course, may 
vary with the individual observer, with the nature of the optical sys- 
tem used for observation, with the color of the light, etc. 

Hence, for comparing the subjective effect with G, graininess rec- 
ords obtained from six widely different types of emulsions are shown, 
each being placed next to its photomicrograph of 300 diameter. This 
comparison requires approximately the same density of the different 

Figs. 5A to 57*" represent thus reproductions of emulsions which are 
typical representatives of negative and positive material used for pro- 
fessional and amateur purposes: 

(a) Material for lithographic reproductions (density: 0.46) G = 

(6) Positive film (density : 0.47) G = 57. 


(c) Sound recording film (density : 0.50) G = 63. 

(d) Process emulsion for purpose of reproduction (density: 0.45) 
G = 59. 

(e) Panchromatic emulsion of medium sensitivity (density: 0.41) 
G = 93. 

(/) Panchromatic motion picture film of very high sensitivity 
(density: 0.47) G = 105. 

The comparison of G-values shows perfect qualitative agreement 
with the subjective experience after which the emulsions are arranged. 
They demonstrate, in addition, a number of rather interesting facts 
at closer inspection. 

From the previous description it is obvious that the smallest detail 
in a graininess record is equivalent to the size of a single grain, and it 
is interesting to find on the records that the shape of the smallest de- 
tail varies considerably with the type of emulsion inasmuch as it is of 
almost equal size in Fig. oA, 5B, 5C, and is considerably larger in the 
last two emulsions of known large granularity (average grain size). 
The first three emulsions show, in spite of an approximate equal grain 
size, considerable variations in the size of the fluctuations, *. e., in 
graininess. Furthermore, it appears that, apart from granularity 
and graininess, each emulsion possesses a rather characteristic shape 
of an average fluctuation which indicates that in different emulsions 
the grains group themselves in a more or less typical fashion. The 
fluctuations of all emulsions were found to obey the probability law 
sufficiently well so that the G- value is descriptive of the occurrence of 
small as well as large fluctuations. 

Graininess and Density. The dependence of graininess on density 
lies in the nature of the former : It must depend upon the density of 
the emulsion because the probability for the occurrence of larger 
fluctuations depends naturally upon the number of grains present so 
that one should expect a larger graininess for a larger photographic 
density of the same emulsion. This dependence of the graininess 
upon the density is demonstrated in Figs. 6^4-6-E, where samples of 
the same (panchromatic) emulsion have been analyzed for five dif- 
ferent densities. The data resulting are the following : 

Fig. 5 Density G 

(A) 0.10 58 

(B) 0.25 75 
(O 0.41 93 

(D) 0.67 92 

(E) 1.09 114 



[J. S. M. p. E. 

Close inspection of the records reveals that the width of the smallest 
detail of the recorded pattern is (indicative of the size of the individual 
grain) unaffected by the density. This is not true any more for 
high densities (Figs. 62? and .E) where the grains begin to "overlap," 
which fact is indicated by the occurrence of very broad fluctuations 
which then result in the decrease of the slope of the curve. 



FIG. 9. G-D diagram: G\ and d are the graininess- 
density functions of two different emulsions from which 
G,\ and G& have been derived by transformation. 

Fig. 9 represents these results in diagrammatic form where the 
graininess of two different emulsions is plotted (as heavily drawn 
curves) against the density [G = f(D)] . Obviously the functions are 
in qualitative agreement with the subjective measurements. Al- 
though one should expect G = for D = 0, experience shows that 
considerable inhomogeneity remains in this case, partly due to the 
celluloid base and partly to the gelatin and the fog. The graininess 


produced by these factors can be estimated, depending on the con- 
ditions, to be between 10 and 20 so that the effect of base, gelatin, 
and fog can produce easily 30 to 50 per cent of the graininess of fine- 
grain emulsions at low densities. 

In order to render the graininess-density function representative 
of the graininess impression under various conditions, two principally 
different types of graininess realization must be distinguished : First, 
the case of relative transparency fluctuations (AT/T m ) to which G 
refers and which are realized by determining the amplitude and fre- 
quency of the fluctuations with constant field brightness (constant 
transmitted light). Second, the case in which the absolute transpar- 
ency fluctuations (AT") are realized. In this case the function G = 
f(D) is not directly representative of the graininess impression as the 
transparency fluctuations are obtained by constant illumination 
(constant incident light). The graininess realized under these con- 
ditions will be called G s as it is closely analogous to the graininess 
effect in sound reproduction. 

Although it would be possible to use the graininess meter under 
the latter conditions it is not necessary to do so for the determination 
of G s because the mathematical definition of G permits, by simple 
calculation, the transformation of G = f(D) into G s = f s (D}. Since 
the probability function used has the simple relation between G and 
( &T/T m ) a , the average relative fluctuation : 

G = (A77r m ) a -\A- or G-T m /\fc = (Ar) a 
G s is then defined in analogy to G as : 

G, = (ATVV^ or G, = G-T m = G-W~D 

With the aid of a transparency-density table it is thus easy to reduce 
a known graininess-density relation into G s = f s (D). It is obvious 
that the two functions are considerably different; the difference is 
the larger, the larger the density; and the origin of both functions is 
the same as G = G; for T 1 (D 0) . 

This transformation from G to G s results in the curves in Fig. 9. 
The G s function differs mainly from the former by a maximum* at 
fairly low densities after which G s declines and approaches zero for 
D = oo . The fact that in Fig. 9 the numerical values of G s are 

* The density at which this maximum occurs depends, of course, on the shape 
of the G function. 



[J. S. M. P. E. 

smaller than G should not lead to the erroneous impression that the 
disturbing effect of the graininess, if realized as G s , is relatively smaller 
than if realized under the conditions of G, because the subjective im- 
pressions received through either eye or ear are not commensurable 
in any case. The absolute values occurring are, of course, correctly 
presented by G s as well as by G. 

The Effect of Scattering on Graininess Measurements. The depen- 
dence of the graininess on density renders the scattering effect of the 
emulsion important. As is well known, the spatial distribution of 

FIG. 10. A schematic demonstration of the influence of scattering upon 
observations with two different apertures: (7) the situation arising for a 
perfectly scattering emulsion; (//) for an imperfectly scattering emulsion. 

the illuminating light is changed as it is diffracted and scattered on 
the grains while penetrating the emulsion (Callier effect). The re- 
sulting change of the light distribution is largest for parallel and least 
for scattered illumination for a given sample. The absolute inten- 
sity at which the fluctuations (AT") are realized depends, conse- 
quently, on the aperture of the receiving system in a non -proportional 
manner, as illustrated in Fig. 10, / and II. The former demonstrates 
the spatial light distribution in polar coordinates over an emulsion 
element illuminated with parallel light, if the former is a perfect 


scatterer. The left half of / (Fig. 10) represents the conditions for a 
transparency twice that of the right half, as the areas of the circles 
which represent the total light intensity transmitted through the 
emulsion are related by \/Az/A\* = l /2orAi/A 2 = -^4. If this light 
is received by two optical systems of which one has the small aperture 
/, the other the large aperture F, the intensity observed through the 
former will be obviously smaller than in the latter, but the variation 
of transparency observed will in both cases be the same since the 
spatial distribution is the same. If the emulsion would not scatter, 
the difference between the intensities found at/ and F would be very 
small; the transparency ratio observed would, however, remain the 

If the emulsion scatters only part of the transmitted light the situa- 
tion shown in Fig. 10 (//) occurs, where the left half demonstrates an 
emulsion of larger scattering qualities than the right half, the diffuse 
transparency being approximately the same for both cases. It is 
evident that the increase of intensity when increasing/ to Fis different 
from the former case (/) and, furthermore, differs between the left 
and the right half of (IT). Consequently a photometric transparency 
measurement for a given (small) aperture can be made only for similar 
spatial distributions. This restriction can introduce a serious error 
in graininess measurements, as the spatial distribution depends not 
only on the grain size but also on the number of grains per unit area, 
i. e., the transparency itself, so that in order to determine &T/T m the 
same aperture has to be used for measuring T m as well as AT", the 
absolute size of which is irrelevant (as long as sufficient resolving 
power and light-intensity are obtained). It is thus incorrect to 
measure, e. g., the total light transmitted by a sample with a photocell 
at small distance (large aperture) and observe or record through an 
optical system at relatively small aperture. In this case the graini- 
ness values, if measured in terms of relative fluctuations, will appear 
too small, the more so the larger the scattering power of the emulsion. 

The above description of the graininess photometer demonstrates 
that this difficulty is avoided inasmuch as none of the diaphragms in 
the scanning system decrease the aperture of it below that of the inte- 
grating system. Nevertheless, an effect occurs which is in certain 
respects analogous to the consequences of unequal apertures, for the 
experience with the graininess photometer has proved that the average 
amount of light received at the scanning area a (see above) and that 
at the integrating cell (Mil, Fig. 1) is not proportional for different 

298 GOETZ, GOULD, AND DEMBER [j. s. M. P. E. 

samples if the scattering is appreciable, in spite of the same aperture 
of both systems. In other words, whenever the emulsion is scatter- 
ing, a situation similar to Fig. 10 (77) arises and the average trans- 
parencies measured over the large and the small field are not com- 

The reason for this is probably to be found in the dissimilarity of 
the two light paths so that the illumination of the scanning cell is 
more dependent upon the parallel component of the light transmitted 
through the emulsion than the illumination of the integrating cell. 
As the two cells are balanced with each other without the interposi- 
tion of a scatterer, a change of this distribution affects the light shares 
received by each of the systems. If thus the field brightness B in 
the integrating cell is kept constant (B = 1) the scanning cell re- 
ceives an average illumination b = B and the mean transparency 
measured here is consequently t m ^ T m . The graininess amplitudes 
recorded thus are &tT m ; if, on the other hand, the light-intensity 
is adjusted to a constant average brightness in the scanning system, 
the integrating cell receives, instead, more light and the amplitudes 
are measured in (&t/T m )(B + 8B) = A/// m . 

The G evaluated under the latter conditions can be used directly; 
it represents the graininess measured in terms of the light reaching 
the scanning cell (i. e., the more parallel light components). Cor- 
respondingly, this G-value can be reduced to the T m value of the in- 
tegrating cell as G' = G/(B + 8B} m . G' represents the more parallel 
transparency fluctuations in terms of a wider range of angular light 
components. Consequently G' ^ G. It is not simple to interpret 
the exact relationship between the two quantities. The difference be- 
tween the density functions of G and G' for a panchromatic emulsion 
is demonstrated in Fig. 11. In all other diagrams of this paper, G 
has been chosen as being probably more representative for most pur- 

This scattering effect appears to be independent of the graininess 
and dependent principally on grain size and grain number, and is not 
simply related to the measurement of the Callier quotient Q. It will 
be discussed in a separate paper. 

Subjective and Objective Graininess. In the introductory chapter 
it was stated that the graininess coefficient G is measured in terms of 

* In our previous papers G' has been employed in the diagrams because the 
former instrument did not permit an exact comparison between /,,. and T m . 


the relative transparency fluctuations &T/T m in order to approach 
the logarithmic relation between stimulus and visual sensation (gen- 
erally known as Fechner's law). One should thus expect that the 
graininess density function G = f(D) should represent the subjective 
visual impression under all conditions of illumination of the emul- 
sion; i. e., it should be independent of the absolute value of the field 
brightness as well as of its variation with D. 

It is, however, well known that the logarithmic relationship holds 








FIG. 11. G-D diagram: The two curves demonstrate 
the deviations in G observed on a scattering emulsion 
for higher densities. 

only within a certain range of illumination, below and above which 
range considerable deviation occurs for the normal eye. As the 
brightness of the projection of an emulsion observed visually is fre- 
quently considerably below this range, particularly as far as regions of 
higher densities are concerned, it is unavoidable that deviations of the 
subjective visual graininess realization from the G-D function result. 
Furthermore, this deviation causes a dependence of the graininess 
impression upon the absolute brightness at which the impression 



[J. S. M. P. E. 

takes place, a fact which would not occur if the logarithmic sensitivity 
of the eye would hold over the full range of illumination used in 
photography. Especially the latter consequence is easily realized if 
one compares the graininess of the same emulsion at low and at high 
brightness. Due to the higher capacity for discrimination for 
medium brightness values, the graininess impression is considerably 
larger in the latter than in the former case. It is thus necessary to 
apply certain corrections to the G-D function if the graininess is to 
be realized at low illuminations. These corrections can, however, be 






FIG. 12. 6-B diagram : The discrimination factor 5 is plotted against the 
absolute brightness. The scale for 5 is shown arbitrarily so that 5 = 1 for 
the brightness region in which Fechner's law is valid. 

applied for every case in fair approximation by the following con- 

The difference in brightness AS of two adjacent or alternating 
fields of view necessary to distinguish them visually depends upon the 
average brightness B at which the experiment is conducted. Ac- 
cording to Fechner's law the relative change in brightness (AB/.B) 
should be the same for all light intensities. However, even under the 
most favorable conditions this is strictly true only in a narrow range 
of field brightness (10 to 100 millilamberts). 2 At these intensities 
the value of &B/B proves to be a minimum; therefore the visual 


contrast sensitivity B/ AB is higher in this range than at a lower or 
higher field brightness. Fig. 12 shows the contrast sensitivity or 
discrimination factor 5 as a function of the field brightness 5 = /(log 
B) for a range sufficient to cover all actual conditions under which the 
graininess of emulsions may be realized. It is obviously convenient 
to define the discrimination factor in the brightness range for which 
Fechner's law (db/dB = 0) is valid as 6 = 1. Consequently all other 
5- values must be < 1 , ( 1 6) , thus being always representative of the 
loss of visual discriminating power. 

To obtain the values of the visual graininess G v for a given bright- 
ness B, defined by the field brightness B for D = as B = B Q 10~ D , 
each G-value must be multiplied with the 5-value corresponding to 
the actual field brightness : 

G, = G-()a 

This transformation is shown in Fig. 13, the upper section of which 
describes the dependence of 5 on D for constant illumination at four 
different intensities (B = 1, 10, 100, 1000 millilamberts). It is seen 
that a somewhat involved relation results inasmuch as 8 decreases with 
D for low intensities, remains practically constant (5=1) for B Q = 
100 millilamberts and increases for extremely large values of B . The 
reason for the latter is due to a "blinding" effect (glare) occurring at 
low densities. 

In the lower section of Fig. 13 the Gi-D function of Fig. 9 is shown 
dashed, and each G-value has been multiplied with the 6-values for 
the corresponding D, resulting accordingly in four G v functions which 
represent the visual graininess impression of the same emulsion for the 
four different intensities of (constant) illumination given above. 

It follows from the previous discussion that the G-D function rep- 
resents the upper limit of all G v values. Whether the deviation of 
G v from G occurs at low or at high densities is determined only by the 
intensity of illumination. For extremely low levels of brightness the 
G v function does not even approach the G-function. Moreover, the 
transformation causes a change of the shape, for depending on 
the shape of the particular 8-D function used a maximum can occur 
for G V (D}, evident in Fig. ll,for^ = 1 and 10 millilamberts. 

The general shape of 5(Z>) determines, of course, the deviation of 
G v from G. Although the seemingly best available values have been 
used in Figs. 12 and 13, it is not quite certain whether the function 
represents perfectly the physiological process involved in the realiza- 



U. S. M. P. E. 

tion of the graininess, since the shape of test-objects used differed 
from general graininess textures. Fundamental deviations from 
the function used are not to be expected. 
Concerning the acoustic realization of graininess by the noise 

FIG. 13. The effect of field brightness upon the 
visually observed graininess. In the upper section 5 is 
plotted against D for B = 1 10 s millilambert. In 
the lower section the G\ curve of Fig. 9 (dashed) is trans- 
formed with the above 5(Z>) into subjective graininess- 
density functions for different illuminations. 

level, similar considerations must be made : as long as the noise level is 
measured with an instrument in terms of energy the G s curve derived 
in Fig. 9 is representative, provided that all elements in the sound 
equipment are linear. If the subjective noise impression is to be rep- 
resented, the G S -D function has to be reduced in the same manner as 


shown above for the visual impression, by replacing 6 with the factor 
of acoustic discrimination (5 S ) and B with the sound intensity. The 
reason why, in the case of visual observation, the relative transparency 
fluctuations have to be considered, whereas the absolute fluctuations 
are representative of the graininess realized as noise, can be given as 
follows: For visual realization the eye adapts itself to the average 
brightness of the field (~T m ) and observes the fluctuations (^A 7') 
resulting in a realization in terms of 5- AT" '/T m . For sound reproduc- 
tions, however, only the alternating component (~(Ar) ) of the 
light falling upon the photocell of the amplifier will be transformed 
into sound energy, independent of the average light intensity (~ T m ) . 
Consequently the acoustic discrimination factor 5,, will determine the 
relation between the objectively measured sound energy and its sub- 
jective realization [~8 S ( AT)J . 

Comparison with Graininess Measurements of Other Observers. 
Since the main practical purpose of graininess determination lies in 
the evaluation of a defined coefficient which is representative of the 
subjective impression caused by the graininess, it has been shown 
above what factors enter into the interpretation of G in order to 
render it descriptive under different conditions of observation. It 
is thus not out of place to compare the G-D functions obtained by 
other authors with those described above. 

The graininess measurements so far available can be divided into 4 
classes : 

(1) Subjective visual comparison with standards. (Jones and Deisch, 4 Hardy 
and Jones, 6 Crab tree, 6 Lowry, 7 Conklin. 8 ) 

(2) Objective photometer methods, (van Kreveld and Scheffer, 9 Siedentopf, 10 
Selwyn. 11 ) 

(3) Noise level. (Narath. 12 ) 

(4) Caillier coefficient. (Threadgold, 13 Eggert and collaborators, 14 Narath. 12 ) 

The instrument developed by van Kreveld and Scheffer 9 is in 
principle a microphotometer connected with an automatic device 
which evaluates its indications in terms of the average density fluc- 
tuations. To allow a comparison of their results, given for a nu- 
merically different unit for the graininess constant, two typical repre- 
sentatives of their graininess-density curves were reduced (Fig. 14) 
so as to coincide at D = 0.5 with the curve G\ of Fig. 9. The same 
reduction was applied to the curve GZ, thus permitting a fair compari- 
son of the shapes. The agreement is obviously satisfactory for lower 



[J. S. M. P. E. 

densities whereas at large densities van Kreveld's curves show flat 

Attempts at mathematical analysis of the density fluctuations to 
be expected in a photographic emulsion and its dependence on the 
scanning area and grain size were given by Selwyn 11 and Siedentopf . 10 
Assuming a Gaussian distribution for the density fluctuations, both 
authors predict a graininess-density dependence : G ~ d\/D accord- 



FIG. 14. Comparison of graininess-density functions 
obtained by different methods and observers. For 
commensurability all curves have been reduced to an 
identical G-value for D = 0.5. The dashed curve (S) 
represents the parabola G ~ dy/ D; K\, Kt are by van 
Kreveld; 9 G\, G* are identical with those in Fig. 9. 

ing to which the G-D function should be parabolic for a constant 
average grain diameter d. (The proportionality factor depends on 
the units for G and the size of the scanning area.) Both authors prove 
by experimental data for different types of emulsions that the as- 
sumptions upon which the calculations were based are valid for the 
usual photographic emulsions. This parabola, coinciding at D = 0.5 
with the experimental curves, is shown dashed in Fig. 14. Again 
the agreement with G\ and Gj is good for low and medium densities ; 


all experimental curves show, however, lower graininess values than 
predicted above D = 1. This deviation between all experimental 
curves and the parabola for high densities may have several reasons 
which tend to decrease the graininess for the presence of many grains ; 
aside from this, the decrease of grain size with the density, as shown by 
Eggert and Kuester 14 and not considered in the relation, may cause 
an effect in this direction. 

It was stated in a previous section that for sound reproduction the 
average absolute transparency fluctuation measured by G s will be 
indicative of the background noise. Narath finds a maximum of the 
noise level as function of density for D = 0.2, which is in good accord- 
ance with the position of the maxima of the two G^-curves derived 
from the original G-curves obtained with the graininess meter. 

As far as the results of the subjective methods are concerned, an 
instructive survey is given by Lowry, whose studies on the influence 
of the field brightness on the visual graininess impression are of 
special interest. The method consists in observing a virtual mag- 
nified image of the emulsion and measuring the magnification at 
which the deposit ceases to appear inhomogeneous. A halftone 
screen serves as standard. The observations of the graininess-den- 
sity dependence at different illumination levels (B = 100, 50, 15, 
and 6 millilamberts) allow a comparison with the results obtained 
from our graininess measurements and corrected according to the 
visual discrimination curve to represent the subjective impression at 
varying field brightnesses (Fig. 13). The general dependence of the 
graininess realization on density and illumination is similar, though a 
larger decrease at high densities is observed, resulting in a sharper 
maximum toward lower values of D, this effect being more pro- 
nounced at low light intensities. This can be explained by the de- 
pendence of the S-B function (Fig. 12) upon the nature of the test- 
object, as pointed out before, since as Hansen and Keck 15 have 
stated the discriminating capacity of the eye is not only determined 
by illumination and contrast, but also by the texture of the object. 
In this respect Conklin's 8 comparison of the graininess of different 
emulsions with each other is probably the closest approach to a cor- 
rect situation in this particular sense. 

Graininess and Grain Size. An entirely different principle is in- 
volved in the graininess determination by means of the Callier effect. 
Threadgold 13 suggested first the use of the Callier quotient Q = 
D s /D d = specular density/diffuse density, as a quantitative repre- 


sentation of the "graininess" ; later Eggert and his collaborators 14 
proposed the use of the coefficient K = 100 logio Q for D = 0.5 for 
the same purpose. By means of a specially designed instrument, 
"the granulometer," K has been very carefully determined for a 
large number of emulsions in comparison with microscopic determi- 
nations of the average grain diameter d. A linear proportionality 
between d and K was found. 

If according to these measurements it is taken for granted that 
this semi -empirical relation is truly representative of the grain diam- 
eter under all conditions, still the question arises whether the grain 
diameter determines the graininess. Obviously this is not true in 
the case of the positive print, as the fluctuations occurring here are 
largely determined by the graininess of the negative superimposed 
upon the graininess of the positive emulsion, without affecting the 
grain size of the latter. Experiments prove that the scattering power 
of a positive emulsion remains unchanged regardless of whether it 
was exposed directly or through a negative. 

Since the practical interest centers chiefly upon the fluctuations 
occurring and not upon the grain size itself, and as, furthermore, the 
positive copy represents in the majority of cases the final photo- 
graphic product and the actual manifestation of the graininess, mea- 
surements of this type can not replace, in general, direct determina- 
tions of fluctuation frequencies. 

Less obvious is the problem when only "primary" graininess is 
concerned. If the graininess G represented by the frequency of trans- 
parency fluctuations could be determined by the average size d of the 
grain whose position relative to its neighbors fluctuates, and if one 
assumes the general validity of a Gaussian distribution for all grain 
arrangements occurring in emulsions, the proportionality G~d 
follows only if one assumes in addition the constancy of the degree 
of disturbance represented by all the influences which contribute to 
the irregularity of the final grain arrangement.* 

* An analogy with Gallon's board may clarify this point: If a large number of 
equal sized balls, starting from one point, roll over an almost vertical plane into 
equidistant boxes at the lower end while the path is obstructed by pins arranged 
in a certain pattern, the distribution of the balls in the boxes will follow the proba- 
bility function e~^ x ^ where I/A (= G) determines the degree of average dis- 
turbance the balls experienced on the way to the boxes. It results in a distri- 
bution which is the broader the larger the average disturbance ; the latter is deter- 
mined by the arrangement of the pins and the size of the balls. The disturbing 


To assume that this average disturbance of the grain is the same 
for all emulsions is not obvious, even though it may very well be 
similar for similar emulsions and similar processing. 

These considerations lead to the conclusion that the use of the 
Callier quotient facilitates a simple and valuable method for the 
measurement of the grain size, but that it does not lead to a direct 
relationship with the actual graininess. 

Graininess and Grain Distribution. The problem of the type of 
distribution law valid for the grain arrangement in an emulsion has 
been discussed in detail previously. 1 Although there does not ap- 
pear an a priori reason for the necessary validity of the probability 
law and no reliable, truly statistical method has been developed to 
prove this, it has generally been assumed to be true. The G-deter- 
mination with the graininess meter is, to our knowledge, the only 
method which permits a precise check on the validity, for not only 
are a very large number of fluctuations considered in the measure- 
ment, but also the frequency for each size of fluctuation is determined 
in terms of the Gaussian function (by comparison of the path of the 
light-beam with a G-spiral of the integrator). 

It is interesting to note that the large number of measurements on 
very different emulsions have shown that in many cases the prob- 
ability function is an astonishingly correct description of the relation 
between frequency and size of fluctuations. There appear, however, 
to occur systematic deviations in emulsions under certain conditions, 
about which we intend to report separately. 

The relation is generally true also for positive prints from negatives, 
as is to be expected, for the superposition of two (or more) truly 

factor remains the same for any size of the board, balls, and pin arrangement as 
long as the proportions remain the same. If only the size of the balls is changed, 
their disturbance by the (now relatively narrower) pin arrangement is increased, 
resulting in a broader final distribution. Since each box determines a certain 
magnitude of deviation x, from the average (center of the board) and the number 
of balls within the frequency of this deviation, and as the ball size stands for the 
grain diameter, the width (~l/&) of the distribution in the boxes represents the 
graininess. A proportionality between G and d means the assumption of either 
an invariable disturbance with a linear proportionality between ball size and width 
of the distribution or a variation of the disturbance with the ball size in such 
manner that the simple above relationship occurs. The disturbance in an emul- 
sion must be interpreted as the sum of all physico-chemical effects in the manu- 
facturing, exposing, and processing of an emulsion which determines the grain 


statistical distributions will again produce a similar distribution of a 
disturbance greater than that of either component. To what ex- 
tent, however, this relation is quantitatively fulfilled for positives is 
not yet sufficiently investigated. 

If one could take the strict validity of the probability relation for 
granted in all cases, it would be necessary only to measure an average 
deviation of the transparency from the mean for the determination 
of the graininess, as several authors have done, for a known distribu- 
tion function permits the calculation of the frequency of any size of 
deviation from a known average size. The experience, however, 
that one can not rely upon the strict validity of the Gaussian dis- 
tribution in an emulsion renders a type of measurement necessary 
which "weighs" each size of deviation against all others. There is 
thus the possibility of determining the frequency of fluctuations, the 
size of which may be particularly important for a given application 
of the emulsion and one can, furthermore, eliminate the effect of 
nonstatistical fluctuations, e. g., scratches, dust particles, etc., which 
occur in the measurement as nonsystematic deviations from the G- 


The establishment of an absolute and objective graininess coeffi- 
cient and the construction of an instrument which permits the mea- 
surement of this coefficient in a comparatively simple manner, 
renders possible the prediction of the graininess realization under 
different conditions of observation. Only two of these different types 
have been discussed the deviation of the noise levels from the graini- 
ness density function, and the corrections for known deviations from 
Fechner's law with regard to the discrimination coefficient of the eye. 

The problems are, of course, not the only important ones, as, for 
instance, the question to what extent the graininess of a negative de- 
termines the graininess of a positive made from it and the dependence 
on the methods of printing the nature of the dependence of the 
graininess of an emulsion on its gamma the effect of the so-called 
fine-grain developers upon the relation between grain size and graini- 
ness are questions of no lesser practical importance than those de- 
scribed. We hope to be able to report soon on some of these questions. 

In conclusion, it may be noted that progress along these lines could 
be furthered considerably and the results of different laboratories 
would be rendered comparable, once a general agreement concerning 


the means and units for determining and describing this evasive 
quality of emulsions will be attained. 

Acknowledgments. The graininess meter was built with the aid of 
the Agfa Ansco Research Fund. The authors are indebted also to 
the directors and the staff of the Agfa Ansco Corporation, Bingham- 
ton, N. Y., and the Agfa Raw Film Corporation, Los Angeles, for 
technical cooperation; to C. Zeiss, Inc., for the loan of optical equip- 
ment for tests; to the RCA Manufacturing Company for the dona- 
tion of two recording systems; and last but not least, to the mechani- 
cal skill of Mr. Charles Edler of this laboratory, without which the 
precision required in the instrument could not have been attained. 


1 GOETZ, A., AND GOULD, W. O. : "The Objective Quantitative Determination 
of the Graininess of Photographic Emulsions," /. Soc. Mot. Pict. Eng., XXIX 
(Nov., 1937), p. 510. 

See also: 

DEMBER, A., GOETZ, A., AND GOULD, W. O. : "The Objective Quantitative 
Definition of the Graininess of Photographic Emulsions," Phys. Rev., 54 (1938), 
p. 240. 

GOULD, W. O., GOETZ, A., AND DEMBER, A.: "An Instrument for the Ob- 
jective and Quantitative Determination of Photographic Graininess," Phys. Rev., 
54 (1938), p. 240. 

GOETZ, A., GOULD, W. O., AND DEMBER, A.: "An Instrument for the Absolute 
Measurement of the Graininess of Photographic Emulsions," /. Soc. Mot. Pict. 
Eng. XXXIII (Oct., 1939), p. 469. 

GOETZ, A.: "Graininess of Photographic Emulsions," Photo Technique 1, 
(Jan., 1939), p. 21. 

GOULD, W. O., AND GOETZ, A.: "About the Physical Nature of the Graininess 
of Photographic Emulsions," Phys. Rev., 56 (1939), p. 850. 

2 NUTTING, P. G. : "1919 Report of Standards Committee on Visual Sensi- 
tometry," J. Opt. Soc. Amer., 4 (1920), p. 55. 

3 TUTTLE, C. : "Densitometry and Photographic Printing. Illumination of 
the Negative and Its Effect upon Density," J. Opt. Soc. Amer., 24 (1934), p. 272. 

* JONES, L. A., AND DEISCH, N.: "The Measurement of Graininess in Photo- 
graphic Deposits," /. Frank. Inst., 190 (1920), p. 657. 

5 HARDY, A. C., AND JONES, L. A.: "Graininess in Motion Picture Negatives 
and Positives," Trans. Soc. Mot. Pict. Eng., No. 14 (1922), p. 107. 

6 CRABTREE, J. I.: "Graininess of Motion Picture Film," Trans. Soc. Mot. 
Pict. Eng., No. 11 (1927), p. 77. 

7 LOWRY, E. M.: "An Instrument for the Measurement of the Graininess of 
Photographic Materials," /. Opt. Soc. Amer., 26 (1936), p. 65. 

8 CONKLIN, O. E.: "Some Applications of the Comparison Microscope in 
the Film Industry," /. Soc. Mot. Pict. Eng., XVI (Feb., 1931), p. 159. 

9 VAN KREVELD, A.: "Objective Measurements of Graininess of Photographic 
Materials," J. Opt. Soc. Amer., 26 (1936), p. 170. 


VAN KREVELD, A., AND SCHEFFER, J. C.: "Graininess of Photographic Ma- 
terials in Objective Absolute Measure," J. Opt. Soc. Amer., 27 (1937), p. 100. 

10 SIEDENTOPF, H.: "Ueber Kornigkeit, Dichteschwankungen und Vergros- 
serungsfahigkeit photographischer Negative," Phys. Zeitschr., 38 (1937), p. 

11 SELWYN, E. W. H.: "A Theory of Graininess," Phot. J., 75 (1935), p. 571. 
SELWYN, E. W. H.: "Experiments on the Nature of Graininess," Phot. J., 

79 (1939), p. 513. 

11 NARATH, A.: "Ueber die Kornigkeit photographischer Schichten," Kino- 
technik, 16 (1934), pp. 255, 287. 

13 THREADGOLD, S. D.: "The Measurement of Graininess," Phot. J., 72 
(1932), p. 348. 

THREADGOLD, S. D.: "The Callier Coefficient and Its Relation to Graininess," 
Phot. J., 79 (1939), p. 524. 

14 KtJSTER, A.: "Ueber die Kornigkeit und das Auflosungsvermogen photo- 
graphischer Schichten," Veroff. Agfa, 3 (1933), p. 93. 

EGGERT, J., AND KtJSTER, A.: "Callicrquotient and mittlerer Korndurchmes- 
ser entwickelter photographischer Schichten," Veroff. Agfa, 4 (1935), p. 49. 

BRANDES, H.: "Apparate zur Messung der Kornigkeit entwickelter photo- 
graphischer Schichten," Veroff. Agfa, 4 (1935), p. 58. 

EGGERT, J., AND K(:STER, A.: "Statistische Versuche zur Beziehung zwischen 
Callier-Quotient, Kornigkeit und Grenzvergrosserung," Veroff. Agfa, 6 (1939), 
p. 186. 

16 HANSEN, G., AND KECK, P. H.: "Messungen der Kornigkeit an Negativ- 
schichten," Zeitschr. f. wiss. Phot., 37 (1938), pp. 86, 99. 






(These regulations are published herein for information and ready reference by the 
readers of the Journal. Publication in the Journal is therefore in no way related to 
approval or disapproval of these regulations by the Society, although much of the ma- 
terial relating to motion picture projection rooms has been based upon recommenda- 
tions made by the Sub- Committee on Fire Hazards of the SMPE Projection Practice 
Committee, published in the November, 1938, issue of the Journal. 

The following regulations are reprinted from NBFU Pamphlet No. 40, July 1, 
1939, copies of which may be obtained from the National Board of Fire Underwriters, 
New York, N. Y.) 

(1) Application of Rules 

(2) Scope of Regulations 

(3) Arrangement of Regulations 

(4) Approval of Plans 

(5) Definitions 

Part I. General Provisions Regarding the Storage and Handling of Film 

(11) Construction and Arrangement of Buildings 

(12) Electrical Equipment 

(13) Heating Equipment 

(14) Sprinklers and Other Fire Protection Appliances 

(15) Storage of Film 

(16) Film Cabinets 

(17) Film Vaults 

(18) Handling of Film 

(19) Motion Picture Projection and Special Processes 

Part II. Special Provisions for Special Occupancies 

(21) Motion Picture Theaters and Other Occupancies in which the Principal 
Use of Film Is in Motion Picture Projection 



(22) Motion Picture Film Exchanges 

(23) Motion Picture Film Laboratories 

(24) Motion Picture Studios 

(1) Application of Rules. These regulations are intended to 
apply to the storage and handling of nitrocellulose motion picture 
film, in all places except establishments manufacturing such film and 
storage incident thereto. They are not intended to apply to the 
storage and handling of film having a cellulose acetate or other ap- 
proved slow-burning base nor to photographic and X-ray film. (See 
separate regulations on Photographic and X-ray Film.) 

(2) Scope of Regulations. (a) These regulations are intended to 
provide reasonable provisions for the storage and handling of motion 
picture film, based on minimum requirements for safety to life and 
property from fire. 

(6) It is strongly recommended that film exchanges, laboratories, 
and studios be permitted only in sprinklered buildings of fireproof* 
construction. In buildings of non-fireproof construction which have 
been adapted to such occupancies, automatic sprinklers should be 
installed as hereinafter specified and suitable fire cut-offs provided 
between each room in which film is handled or stored and other sec- 
tions of the building, and adequate exit facilities provided. Suitable 
requirements will be found in the Recommended Building Code of the 
National Board of Fire Underwriters. 

(3) Arrangement of Regulations. (a) These regulations are divided 
into two parts : Part I gives general provisions regarding the storage 
and handling of film; Part II gives special provisions for special occu- 
pancies as motion picture theaters, exchanges, laboratories, and 
studios, which apply in addition to any and all of the general pro- 
visions which may also be applicable. 

(b) The grouping of the special provisions under the heading of 
special occupancies is merely for convenience in the application of 
these regulations. Any particular process or operation in any type 
of occupancy shall be governed by the provisions given for that 
process or operation, whether under the heading of that occupancy 
or any other heading, unless otherwise specifically provided herein. 

* The term "fireproof" is used as defined in the Building Code of the National 
Board of Fire Underwriters and as having in these regulations the same meaning 
as the term "fire-resistive" as used by the National Fire Protection Association. 


For example, any process in a studio which, from the standpoint of the 
authority enforcing these regulations, partakes of the same nature 
as some process covered under laboratories, shall be governed by the 
provisions for that process given under laboratories. 

(4) Approval of Plans. Before constructing any building for use as 
a motion picture film occupancy, or remodeling any building for such 
occupancy, or building any film vault, or installing any enclosure for 
motion picture projection, or installing any screening room, complete 
plans of such proposed construction or installation should be sub- 
mitted to the inspection department having jurisdiction for approval. 
These plans shall show in detail all proposed construction and struc- 
tural changes and the means of protection to be provided, the heating 
system with the protection for it, the electrical equipment, and the 
character and location of exposures. 

(5) Definitions. Whenever used in these regulations the following 
words shall be construed as having the meanings given below. 

(a) "Film" or "motion picture film," motion picture or sound 
recording film having a nitrocellulose base, whether in the form of 
unexposed film, positives, negatives, scrap, or used film. 

(b) "Vault," a vault constructed and equipped in accordance with 
the requirements of Section 17. 

(c) "Cabinet," a cabinet constructed and equipped in accordance 
with the requirements of Section 16. 

(d) "Standard roll," a roll of film ! 3 /s inches (35 mm) wide and 
1000 feet long, weighing approximately 5 pounds, used as a unit in 
calculating the weight of film. 

NOTE. This definition is intended to establish a measure of length and weight 
and is not designed to prohibit the use of double rolls (2000 feet) of film in theaters 
and exchanges. 

(e) "Partition," except where some other form of construction is 
specified, a partition constructed in accordance with the specifica- 
tions given in sub-section 112. 


Section 11. -Construction and Arrangement of Buildings 

(111) Motion picture film should preferably be stored or handled 
only in buildings of fireproof construction. 


(112) Partitions. (a) All rooms in which motion picture film is 
stored or handled, except motion picture projection rooms and film 
vaults, shall be separated from each other and from all other parts of 
the building by partitions of suitable stability and having a fire re- 
tardant classification of not less than 1 hour as determined by the 
Standard Fire Test. Partitions constructed as follows shall be 
deemed to have the required fire retardant classification : 

(1) Hollow clay tile laid in cement mortar, cement lime mortar, or 
gypsum mortar, not less than 4 inches thick and plastered on both 
sides with not less than 1 / z inch of gypsum mortar or cement mortar; 

(2) Gypsum blocks, either solid or hollow, laid in gypsum mortar, not 
less than 3 inches thick and plastered on both sides with not less than 
*/2 inch of gypsum mortar ; 

(3) Metal lath supported by incombustible studs, plastered on both 
sides to fully cover the metal lath and studs with not less than 3 / 4 
inch of gypsum mortar or cement mortar and having a total thickness 
of not less than 2 J /2 inches; 

(4) Wood studs covered both sides with metal lath and 3 /< inch gyp- 
sum mortar or cement mortar, and having a total thickness of not 
less than 5 J /4 inches. (This type of construction to be used only 
in buildings not of fireproof construction.) 

(b) Partitions shall be continuous from floor to ceiling and securely 
anchored to walls, floor, and ceiling. 

(c) Openings in partitions shall be protected by approved fire doors 
of a type suitable for use in Class C situations as defined in the Regu- 
lations for the Protection of Openings in Walls and Partitions against 

(113) In buildings not of fireproof construction, all rooms in which 
motion picture film is stored or handled shall have floors and ceilings 
of at least double 7 / 8 -inch tongue and groove boards, or the equivalent. 

(114) Exits. It is essential that all rooms in which film is handled 
be provided with adequate aisle space and safe means of egress. 
Aisle space should not be less than 30 inches clear wherever walking 
is necessary. Rooms in which film is handled and in which more 
than two persons work shall have two or more exits, remote from each 
other. Every exit shall be marked "Exit" in letters not less than 6 
inches high, or by an illuminated sign with letters of the same height. 

(115) Vents. All new buildings erected to be used for film occu- 
pancy, and all existing buildings remodeled for such occupancy, ex- 
cept projection rooms (sub-section 191), rewind rooms (paragraph 


212), and rooms associated therewith (paragraph 213(a)4), shall be 
provided in every room where film is stored or handled, with vents 
that will open automatically in case of fire. These should be of ample 
size; they may be in the form of automatic skylights or automatic- 
opening window sash. All rooms in which film is stored or handled 
in existing buildings shall be provided with such vents wherever prac- 

(116) Spacing of Workers. A feature which often contributes ma- 
terially to the hazard to life in film handling rooms is the congestion 
of workers together with large quantities of film. To prevent such 
congestion of workers and the attendant hazard to life, the number of 
persons working in a room where film is handled should never be 
more than will result in a ratio of floor area to number of workers, 
less than 35 square feet per person. Not over 15 persons shall work 
at one time in any one room (not including the stage of motion pic- 
ture studios) in which film is handled. 

(117) Tables and Racks. Tables and racks used in connection with 
the handling of film (joining, inspection, and assembling tables, for 
example) shall be of metal or other non-combustible material. They 
should be kept at least 4 inches away from any radiator or heating ap- 
paratus. Tables shall not be provided with racks or shelves under- 
neath them, which might be used for keeping film or other materials. 

Section 12. -Electrical Equipment 

(121) Artificial illumination in any room where film is handled or 
stored shall be restricted to incandescent electric lights, except that 
arc lights or other forms of electric lights may be used in studios. 

(122) All electrical wiring and equipment shall conform to the 
National Electrical Code. The wiring method shall be rigid metal 
conduit or other approved type of metal raceway. Fuses shall be 

(123) Light fixtures shall be firmly fixed in place, and lights shall 
be protected by vapor-tight globes. All lights shall be equipped with 
keyless sockets and operated by wall switches. 

(124) Light boxes reading "EXIT" in letters not less than six 
inches in height shall be placed at the exits of all darkrooms. 

(125) Portable electric lights on extension cords shall not be used 
in any room in which film is handled or stored, other than the stage of 
motion picture studios, except that in emergency such portable lamps 
may be used if equipped with approved keyless sockets and metal 



[J. S. M. p. E. 

protective lamp guards, and having rubber-covered cords of the 
Hard Service (type 5) or Junior Hard Service (type SJ) varieties, 
with suitable locking plugs. 

(126) Motors shall be of the non-sparking type, or shall be of an 
enclosed type, so arranged as to minimize the danger of sparks. 

(127) Motion picture projectors and other associated electrical 
equipment shall be of approved type and safeguarded in accordance 
with the requirements of the National Electrical Code. 

FIG. 1. 

FIGS. 1 AND 2. Method of Guarding Radiators. 

Guard of Y^inch mesh galvanized steel wire cloth 20 gauge. Bottom 
hinged to lift up for cleaning purposes. Top slopes so articles placed thereon 
will slide off. The top therefore can not be used as a shelf. 

Section 13. Heating Equipment 

(131) Artificial heating in any building or room, other than a vault, 
in which motion picture film is used, handled, or stored, shall be re- 
stricted to steam not exceeding 15 pounds' pressure or hot water, pro- 
vided, however, that this shall not be construed as prohibiting the in- 
stallation of an indirect system employing high pressure steam when 
the radiators or heating coils of such system are not located in the 
room or rooms to be heated. Heat generating apparatus shall be 
in a separate room. 

NOTE. Ordinary hot air furnaces are prohibited. Gas, oil, and electric 
heaters are prohibited in rooms where film is handled or stored. 

(132) All steam pipes within 6 feet of the floor, and where passing 
through partitions or racks or near woodwork, shall be covered with 

Mar., 1940] 



approved pipe covering. All radiators, heating coils, and pipes and 
returns that are near the floor or are so located as to permit any com- 
bustible material, waste or dirt to come in contact therewith shall be 
guarded and protected by means of Y-rinch mesh galvanized steel 
wire cloth No. 20 B. & S. gauge, or by its equivalent. The bottoms 
of such guards shall be arranged so as to lift up for cleaning purposes 
and the tops to slope so that guards can not be used as shelves. 
Guards shall be so constructed that no film can come within 4 inches 
of the heating surface, and shall be made with a substantial metal 
framework which will prevent the wire mesh being forced against 
the radiator or pipes. 

FIG. 2. 

(133} Air conditioning, warm air heating, air cooling, and ven- 
tilating systems employing ducts shall be installed in accordance with 
the Regulations on Air Conditioning, Warm Air Heating, Air Cooling, 
and Ventilating Systems. In addition to the fire dampers required by 
said regulations, approved automatic fire dampers shall also be lo- 
cated at such points as may be necessary so that, as far as the duct 
system is concerned, each room in which film is handled is cut off by 
dampers from every other room, including those where film is handled 
as well as those where film is not handled. (See Par. 191 (g) regarding 
ventilation of projection rooms.) Any system used for air condi- 
tioning a film vault shall be entirely independent, with no duct con- 
necting to any other vault or room. 


Section 14. Sprinklers and Other Fire Protection Appliances 
NOTE. Sec sub-section 175 regarding sprinklers in film vaults. 

(141) Every room in which film is stored or handled in quantities 
greater than 50 pounds (10 standard rolls), except in motion picture 
projection booths or rooms and rewinding rooms connected therewith, 
shall be equipped with an approved system of automatic sprinklers. 
Buildings or sections of buildings used as exchanges, laboratories, or 
studios shall be equipped with automatic sprinklers, as provided under 
sub-sections 221, 231, and 241. All buildings used for the storage or 
handling of film should be completely equipped with automatic 

(142) The spacing of sprinkler heads in all sections where film is 
handled shall not exceed one head for each 64 square feet, with heads 
and lines not over 8 feet apart; provided, that in the stage section of 
motion picture studios the spacing of sprinklers shall not exceed one 
head for each 80 square feet. In existing buildings where the spacing 
of sprinkler heads exceeds that specified above, the inspection depart- 
ment having jurisdiction may require the installation of additional 
heads wherever the hazard of some machine, process, or accumulation 
of film warrants such protection. 

(143) (a) Water supply shall be provided acceptable to the in- 
spection department having jurisdiction. 

(b) Water supplies for automatic sprinklers shall be based on an 
estimate of 20 gallons a minute per head for 20 minutes for the total 
number of heads in one vault, plus 25% of the number of heads in the 
largest fire area. (A fire area is regarded as an area cut off by brick 
or concrete walls having a minimum thickness of 8 inches. Each 
opening in these walls to be protected by one self-closing fire door, 
Class A type.) 

(144) Every room in which film is stored or handled, except film 
vaults, shall be provided with first aid fire appliances of types using 
water or water solutions. 

NOTE. Small hose equipment is recommended, and the following types of 
extinguishers are considered suitable: soda acid, calcium chloride, pump tank, 
and loaded stream. 

See Regulations on First Aid Fire Appliances, and Standpipe and Hose Systems. 

Section 15. Storage of Film 
(151) The storage of motion picture film, not in process or being 


worked upon, and except as hereinafter specifically provided shall be 
in accordance with the following rules: 

(a) Except as provided in paragraph (b) 

(1) Amounts in excess of 25 pounds (5 standard rolls) but not in ex- 
cess of 1000 pounds (200 standard rolls) shall be kept in approved 
cabinets if not in vaults; 

(2) Amounts in excess of 1000 pounds shall be kept in vaults; 

(3) Storage for any considerable length of time should be in vaults 

(&) Unexposed film enclosed in the original shipping cases, conform- 
ing to I. C. C. regulations with each roll in a separate container, shall 
be kept in a sprinklered room, and if over 5 cases aggregating in excess 
of 750 pounds (150 standard rolls) shall be kept in a sprinklered room 
used for no other purpose. 

(152} Valuable negatives shall be stored in vaults used only for 
such film, in suitable heat-insulating containers designed to minimize 
water damage. 

NOTE. The above paragraph being principally concerned with safeguarding 
values would not need to be inserted in an ordinance. 

Section 16. Film Cabinets 

(161) Construction. (a) Cabinets including doors shall be of a 
type of construction approved by the inspection department having 

NOTE. Cabinets may be of approved metal construction, or may be built into 
the building with a type of construction listed under partitions, sub-section 112 
(a) 1,2, or 3, if otherwise conforming to the provisions of this section. 

(&) Cabinets shall have a capacity of not in excess of 375 pounds of 
film (75 standard rolls). 

(c} Racks in the cabinet shall be of metal and so arranged that 
containers will be stored on edge only. 

(d) Doors shall close tightly against the jambs, and should be so 
arranged as to remain normally closed and latched. 

(162} Vents. (a) Cabinets having a capacity of over 50 pounds of 
film (10 standard rolls) shall be provided with a vent from each 
compartment to the outside of the building. The vent shall have a 
minimum effective sectional area of 14 square inches per 100 pounds 


of film capacity. For long lengths of vent pipe a larger size may be 
necessary to take care of friction loss and turns in the pipe. 

(b) Vent flues shall be of construction equivalent to 18 U. S. gauge 
riveted sheet metal, and where inside the building shall be covered 
with 1 inch of approved heat insulating material. 

(163) Sprinklers. (a) Cabinets holding over 75 pounds of film 
(15 standard rolls) shall be provided with at least one automatic 
sprinkler; provided, however, that a cabinet constructed so that 
each roll is in a separate compartment and will burn out without 
communicating fire to film in any other compartment, need not be 
provided with an automatic sprinkler. 

(6) Cabinets of not over 125 pounds' capacity for use in projection 
booths and rewinding rooms only, may have the required sprinkler 
head connected to the house supply by not less than 3 /4-inch pipe, 
provided the water pressure at that elevation be not less than 15 
pounds, and is sufficient to supply not less than 15 gallons a minute. 

(164) Film in cabinets shall be in individual roll containers or in 
I. C. C. shipping containers. Materials other than film shall not be 
stored in the same cabinet with film. 

Section 17 Film Vaults 

(171) Construction. (a) Vaults shall be constructed in accordance 
with plans submitted to and approved by the inspection department 
having jurisdiction. 

(b) Vaults shall not exceed 750 cubic feet in inside dimensions. 

(c) Walls and floor shall be constructed of not less than 8 inches of 
brick, 6 inches of reinforced concrete, or of 12 inches of hollow tile 
plastered on both sides with cement plaster to a thickness of at least 
l /z inch; they shall be without cracks or holes permitting escape of 
gases of combustion into the building. 

(d) Vaults shall be supported by masonry or steel of sufficient 
strength to carry the load safely. Beams shall rest at both ends on 
steel girders, iron or steel columns, or walls or piers of masonry. 
The supports shall afford at least 4 hours' protection as determined 
by the Standard Fire Test. Hollow tile shall not be used for founda- 
tion walls or for walls of other than the top vault where vaults are 

(e) The roof shall be of reinforced concrete at least 6 inches thick ; 
where the floor or roof above is equivalent to this, it may serve as the 
vault roof; a heavy wire screen of not less than 2-inch mesh, or its 

Mar., 1940] 



equivalent, may be installed below the required roof to limit the in- 
terior vault space to 750 cubic feet. 

(/) Vaults shall be provided with suitable drains or scuppers to the 
outside of the building. 

(g) Proximity to stacks and other sources of heat shall be avoided. 

(172} Doors. Door openings shall be protected with approved fire 
doors, one on each face of the wall. 

NOTE. Vaults may have two door openings. Such an arrangement is often 
a great convenience, as in laboratories, where the vault is located between rooms 
and used for the temporary storage of film in process. 

FIG. 3. Film storage vault. Showing construc- 
tion of racks and installation of partitions, baffles, 
and sprinklers. 

Doors shall be of the type suitable for use in Class B situations as 
defined in the Regulations for the Protection of Openings in Walls and 
Partitions against Fire. The interior door shall be automatic. The 
outer door shall be of the swinging type and close into an approved 
frame or otherwise made tight to prevent the passage of flame around 
the edges. It shall be self-closing, and if fastened open shall be 
arranged to close automatically in case of fire originating in or out of 
the vault. Approved quick-operating devices for closing vault doors 
are recognized as having advantages over the fusible link, and their 
use is recommended. 

(173) Vents. (a) Each vault shall be provided with an indepen- 


dent vent having a minimum effective sectional area of 140 square 
inches per 1000 pounds of film capacity (equivalent to 70 square 
inches per 100 standard rolls). The vent area for a vault of 750 
cubic feet shall be not less than 1400 square inches. 

NOTE. In determining the proper vent opening, allowance must be made for 
the window frame and sash, for the area of the glass is considered the effective sec- 
tional area of the vent opening. 

(b) Vent flues inside the building shall be constructed of 5 inches 
of reinforced concrete or of a construction equivalent to that required 
for smoke chimneys. Exterior flues shall be of a construction equiva- 
lent to that of smoke stacks. 

(c) The outlet of each vent shall be above roof or shall be made to 
face street, court, or other clear opening which will give a distance of at 
least 50 feet to any window or other opening exposed thereby and 
not in the same plane, and a distance of at least 25 feet to any fire 
escape on the same or higher level. 

(d) Vaults, especially those having a vent in the form of a window, 
shall be arranged in some manner which will protect the film in the 
vault against ignition by 

(1) Rays of the sun, whenever the film in the vault is exposed to 
direct rays of the sun entering through the vent. This may be done 
by painting the glass in the vent opening a dark color ; 

(2) Radiated heat entering through the vent opening, as from an 
exposure fire, whenever the vent is severely exposed by buildings or 
storage of combustible material, or by other openings in the same wall. 

NOTE. To effect the above protection, one method which has been used em- 
ploys two baffle walls inside the vault. The baffle wall nearer the vent should 
extend from the ceiling down to within about 3 feet of the floor, and the inner baffle 
wall from the floor up to within about 3 feet of the ceiling. Baffle walls should be 
of substantial construction and should be so spaced and arranged as to afford the 
full required vent area from the film storage space to the outside. 

(e) Each vent shall be protected against the weather by single 
thickness glass (Vie inch thick), in a sash arranged to open auto- 
matically in case of fire by the means of an appproved releasing 
device placed inside the vault. The use of approved quick-operating 
devices is recommended. The area of the glass shall be the effective 
sectional area of the vent opening. No pane of glass shall be smaller 
than 200 square inches. Any protection equivalent to the above may 
be accepted in lieu thereof. 

Mar., 1940] 



(/) A light wire screen not coarser than 1 / 8 inch mesh shall be placed 
in each vent. Bars or screen designed to prevent burglary or injury 
to contents shall not have a mesh of less than 4 inches, shall be located 
inside the light wire screen, and shall give a net opening equal to that 
called for in (a). Bars and screens shall be so arranged as not to 
interfere with the automatic operation of the sash. 

(g) Film vaults shall not be provided with skylights or glass win- 
dows other than as specified for vents. 

(174} Racks. Racks in film vaults shall be of metal or other in- 
combustible material and arranged for the storage of single reel 
containers on edge or for I. C. C. shipping containers. Negatives 
need not be stored on edge. Vertical incombustible partitions 
equivalent in durability and heat insulation to 3 /s-inch hard asbestos 

FIG. 4. 

View of baffles and partitions in connec- 
tion with sprinklers. 

and extending from floor to top of rack shall be provided to divide 
racks into sections not over 3 feet wide and so placed as not to ob- 
struct distribution from sprinkler heads. Racks shall not obstruct 
vent openings. 

(175) Sprinklers. Vaults shall be protected by an approved sys- 
tem of automatic sprinklers, with a ratio of one head to each 62 1 /% 
cubic feet of total vault space. A vault of 750 cubic feet shall have 
12 sprinkler heads. Sprinkler heads shall be arranged to give uni- 
form distribution within the sections formed by the above mentioned 
partitions. They shall be separated by sheet-metal baffles extending 
below the sprinkler deflectors. When an approved automatic sprink- 
ler system with open heads is permitted by the inspection department 
having jurisdiction, the baffles between heads may be omitted. 

(176) Lights. All lights in film vaults shall be at the ceiling and of 


the fixed type, with vapor-proof globes and conduit wiring. All 
switches shall be outside the vault and should be arranged with a 
small pilot light to indicate on outside of vault whether vault lights are 
on or off. 

(177) Heat. Heating, when required to prevent sprinkler pipes 
freezing, shall be by hot water or low pressure steam with automatic 
control limiting steam pressure to 10 pounds and the vault tempera- 
ture to not in excess 70 degrees F. Radiators shall be placed 
at the ceiling, over aisle space with pipes and radiators protected with 
wire guards so arranged that no film can be placed within 12 inches 
of such pipes or radiators. 

(178) All film in vaults shall be in containers, either in single-roll 
containers which shall be kept on edge on racks only, except that 
negatives need not be stored on edge, or in I. C. C. shipping containers 
which may be kept on the floor. Materials other than film and film 
cement shall not be stored in the vault. 

Section 18. Handling of Film 

(181) Film Shall Be in Containers. All film shall be kept in closed 
containers except during the actual time it is being worked upon or 
examined. This is very essential from the standpoint of fire hazard 
and safety to life. I. C. C. shipping containers and individual con- 
tainers for each roll of film with proper corrugations on each side are 

(182) Film shall not be placed or kept under benches, tables, or 
other surfaces which would shield it from the discharge of sprinklers. 

(183) Scrap Film. Scrap film shall be kept separate from waste 
paper and other rubbish, and shall be kept under water at all times. 
It shall be collected from work rooms at least once daily, and removed 
to a room used for no other purpose, where it shall be kept under 
water in steel drums with tight covers. These drums shall be dis- 
posed of at frequent intervals. Discarded film in full or part rolls 
shall be kept in vaults. Scrap film shall not be baled or burned. 

NOTE. Motion picture film in the form of clippings and short lengths is in a 
very hazardous form. Safe precautions in the handling of such scraps are most 
essential. Baling and burning of film are processes offering a distinct fire hazard. 
Sending film to a central reclaiming plant in lieu of burning is recommended. 

(184) Transportation. (a) Motion picture film should never be 
transported in any vehicle or other public conveyance used for the 


transportation of passengers, unless enclosed in I. C. C. shipping con- 

(6) Motion picture film should never be allowed in any under- 
ground subway train or station unless under the jurisdiction of the 
Interstate Commerce Commission and conforming to the regulations 

Section 19. -Motion Picture Projection and Special Processes 

(191) Enclosures for Motion Picture Projectors. (a) Motion pic- 
ture projectors using nitrocellulose film shall be operated or set up for 
operation only within an approved enclosure, not less than 48 square 
feet in size and 7 feet high. If more than one machine is to be oper- 
ated an additional 24 square feet shall be provided for each additional 

For new construction, a size not less than 8 feet wide, 10 feet deep, 
and 8 feet high is recommended for one projection machine, and not 
less than 14 feet wide, 10 feet deep, and 8 feet high for two machines. 

(6) The walls and ceiling of the enclosure shall be built of brick, tile, 
or plaster blocks, plastered on both sides, or of concrete, or a rigid 
metal frame, properly braced, and sheathed and roofed with sheet 
iron of not less than No. 20 U. S. gauge metal, or with y 4 -inch hard 
asbestos board, securely riveted or bolted to the frame, or 2 inches of 
solid metal lath and cement or gypsum plaster. All joints shall be 
sufficiently tight to prevent the discharge of smoke. Non-com- 
bustible acoustical material may be used on ceiling and walls, on top 
of the plaster. 

For new construction, it is recommended that the walls of the en- 
closure be constructed in accordance with the requirements of sub- 
section 112, paragraphs (1), (2), or (5), for partitions, with floor 
and ceiling of equivalent fire resistance. Modern heavy equipment 
may require special attention to floor strength and support. In some 
cases it may be necessary to support the projection room inde- 
pendently of the structure. 

(c) The entrance door into the enclosure shall be at least 2 feet by 
5 feet, of construction equivalent to the sheathing permitted above 
for rigid frame construction, and shall be self-closing, swinging out, 
and shall be kept closed at all times when not used for egress or in- 

For new construction it is recommended that at least two doors 
be provided, each not less than 30 inches wide and 6 feet high. Doors 


should be approved fire doors of a type suitable for use in corridor 
and room partitions (Class C openings as defined in the Regulations on 
Protection of Openings in Watts and Partitions}. Exits should be in 
accordance with requirements of authorities having jurisdiction, 
particularly as to size and location. At least one should be of the 
conventional stairway type, having a suitable landing at the top or 
should open directly onto a corridor. 

(d) Two openings for each motion picture projector shall be pro- 
vided; one for the projectionist's view (observation port) shall be not 
larger than 200 square inches, and the other through which the pic- 
ture is projected (projection port) shall be not larger than 120 square 
inches. Where separate stereopticon, spot, or flood light machines 
are installed in the same enclosure with picture machines, not more 
than one opening for each such machine shall be provided for both 
the operator's view and for the projection of the light, but two or more 
machines may be operated through the same opening; such openings 
shall be as small as practicable and shall be capable of being protected 
by approved automatic shutters. 

(e) Each opening shall be provided with an approved gravity shut- 
ter set into guides not less than one inch at sides and bottom, and 
overlapping the top of the opening by not less than one inch when 
closed. Shutters shall be of not less than 10-gauge iron or its equiva- 
lent, or of Y4-inch hard asbestos board. Guides shall be of not less 
than 10-gauge iron or its equivalent. Shutters shall be suspended, 
arranged, and interconnected so that all openings will close upon the 
operating of some suitable fusible or mechanical releasing device, de- 
signed to operate automatically in case of fire or other contingency re- 
quiring the immediate and complete isolation of the contents of the 
enclosure from other portions of the building. Each shutter shall 
have a fusible link above it, and there shall also be one located over 
each upper projector magazine which, upon operating, will close all 
the shutters. There shall also be provided suitable means for man- 
ually closing all shutters simultaneously from any projector head and 
from a point within the projection room near each exit door. Shutters 
on openings not in use shall be kept closed. 

(/) All shelves, furniture, and fixtures within the enclosure shall be 
constructed of incombustible material. Tables shall conform to para- 
graph 117. No combustible material of any sort whatever shall be 
permitted or allowed to be within such enclosure, except the films used 
in the operation of the machine, and film cement. See Section 214. 


(g) Ventilation shall be provided by one or more mechanical ex- 
haust systems which shall draw air from each arc lamp housing and 
from one or more points near the ceiling. Systems shall exhaust to 
outdoors either directly or through a non-combustible flue used for 
no other purpose. Exhaust capacity shall be not less than 15 nor 
more than 50 cubic feet per minute for each arc lamp plus 200 cubic 
feet per minute for the room itself. Systems shall be controlled 
from within the enclosure and have pilot lights to indicate operation. 
The exhaust system serving the projection room may be extended to 
cover rooms associated therewith such as rewind rooms. No dampers 
shall be installed in such exhaust systems. Ventilation of these 
rooms shall not be connected in any way with ventilating or air con- 
ditioning systems serving other portions of the building. 

(h) Exhaust ducts shall be of non-combustible material, and shall 
either be kept 1 inch from combustible material or covered with 
1 /2-hich of non-combustible heat insulating material. 

(i) Fresh air intakes other than those direct to the open air shall 
be protected by approved fire shutters arranged to operate auto- 
matically with the port shutters. 

(j) Provision shall be made so that the auditorium lights can be 
turned on from inside the projection room and from at least one other 
convenient point in the building. 

NOTE. Automatic sprinklers in projection rooms have been very successful 
controlling fires and reducing losses, and their installation is recommended wher- 
ever practicable. 

(193) Processing of Film. The processing of film, as cleaning, 
polishing, buffing, and other special treatments shall not be done in 
rooms where other operations are performed, except that in motion 
picture theaters, cleaning of film may be done in the rewind room. 
(See paragraph 212.} Special processes for treating film shall be 
provided with such proper safeguards as are necessary for protection 
against the hazards involved. The inspection department having 
jurisdiction shall be consulted in regard to the protection needed. 

(194) Soldering Cases. Soldering cases of film when done in a 
building shall be conducted in a room used for no other purpose. 
Walls of the room shall be constructed of 6-inch hollow tile plastered 
each side to a thickness of Va inch or its equivalent. Area of room 
shall not exceed 60 square feet. Opening to room shall not be from 
another film handling room ; it shall be protected by an approved self- 


closing Class B fire door. Automatic vent shall be provided with a 
ratio of 70 square inches for each 500 pounds of film. Rooms shall 
be equipped with automatic sprinklers with a ratio of one sprinkler for 
each 15 square feet with proper sheet-metal baffles. Quantity of film 
in room shall not exceed one case. 

NOTE. The use of shipping cases with metal linings of the telescope type which 
do not need to be soldered is recommended. 

(195) Silver Reclaiming. The process of reclaiming silver from 
film shall not be carried on in a building with other processes unless 
cut off therefrom by standard fire walls. Such sections shall be 
completely equipped with automatic sprinklers. 

(196) Film Cement. Compounds of collodion, amyl acetate, or 
similarly flammable cements shall not be kept in the rooms where 
they are used, in quantities greater than 1 quart; and such material 
in excess of this quantity shall be kept in a vault. The use of these 
materials in motion picture theaters and other special occupancies is 
covered in sub-section 214. 

(197) Smoking. Smoking, except in rooms especially provided for 
the purpose, should be prohibited in any establishment handling or 
storing film, and conspicuous "No Smoking" signs should be posted 
in prominent places. Matches should not be carried by any em- 


NOTE. It is the intent of the regulations to permit the use of 2000-foot rolls 
of 35-mm film in theaters and exchanges only, when handled and stored as pre- 
scribed. The limitations on quantities permitted are based on weight. 

Section 21 . Motion Picture Theaters and Other Occupancies in Which 
the Principal Use of Film Is in Motion Picture Projection 

(211) Enclosure for Projectors. Motion picture projectors shall be 
installed in an enclosure in accordance with sub-section 191. 

(212) Rewinding. (a) Rewinding of films shall be performed 
either in a special rewind room at an approved location, or in the 
projection room. If done in the projection room, approved enclosed- 
type rewind machines should be used. An approved can for scrap 
film having a self-closing hinged cover shall be provided. 

(b) Rewind rooms shall be at least 80 square feet in area, with 
walls and doors in accordance with the requirements of sub-section 
112 and with ceiling of equivalent fire resistance, and shall have a 


vent to the outside of the building of not less than 27 square inches. 
[See paragraph 191 (g).] Exhaust ducts shall comply with para- 
graph 191 (h). Shelves, furniture, and fixtures shall comply with 
paragraph 191(f). 

(213) Care and Use of Film. Motion picture film used in connec- 
tion with the projection of motion pictures (as in theaters, motion 
picture theaters, screening or projection rooms, sound recording 
studios, and motion picture titling studios) shall be limited and kept 
as follows: 

(a) The quantity of film in any projection room or rewinding room 
not equipped with an approved system of automatic sprinklers shall 
be limited to that given below ; if equipped with an approved system 
of automatic sprinklers, double the quantity specified may be per- 

(1) In a projection room, constructed of brick, hollow tile, concrete, 
or other approved masonry, not exceeding 125 pounds (25,000 feet of 
35-mm film) ; 

(2) In a rewinding room constructed of brick, hollow tile, concrete, 
or other approved masonry, separated from projection room with open- 
ings thereto protected with approved fire doors, not exceeding 125 
pounds (25,000 feet of 35-mm film) ; 

(3) In a projection booth constructed of metal frame covered with 
asbestos board or sheet iron not exceeding 75 pounds (15,000 feet of 
35-mm film) ; 

(4) In a special room constructed and vented as required for rewind- 
ing rooms (see sub-section 212), when approved by the inspection 
department having jurisdiction, not exceeding 125 pounds may be 
kept in lieu of the amount permitted in either the projection room 
or the rewind room. The total quantity in the three rooms shall not 
exceed 250 pounds (50,000 feet of 35-mm film). 

(6) The above quantities of film shall be kept as follows : 

(1) Up to 40 pounds (8000 feet of 35-mm film) of film may be kept 
in Interstate Commerce Commission shipping containers, or approved 
cabinet in each room; 

(2) If the amount of film on hand exceeds 40 pounds, an approved 
cabinet shall be provided, in which the amount of film in excess of 
40 pounds shall be kept. 

(214) No collodion, amyl acetate, or other similar flammable 
cement or liquid in quantities greater than 1 pint shall be kept in the 
projection booth or room or rewind room. 


(215} Splices in film shall be made on mechanical cutting and splic- 
ing machines. See paragraph 212(a) on handling of scrap film. 

(216) Location. The number and location of motion picture pro- 
jection rooms or booths in any non-sprinklered building shall be sub- 
ject to the approval of the inspection department having jurisdiction. 

(217) Operation. Motion picture projectors shall be operated by 
and be in charge of qualified projectionists, who shall not be minors. 

(218) Procedure in Case of Fire. In the event of film fire in a 
projector or elsewhere in a projection or rewind room, the projection- 
ist should immediately shut down the projection machine and arc 
lamps, operate the shutter release at the nearest point to him, turn 
on the auditorium lights, leave the projection room, and notify the 
manager of the theater or building. 

Section 22. Motion Picture Film Exchanges 
(See Part I, General Provisions, which also apply.) 

(221) Sprinkler Protection. Buildings not of fireproof construc- 
tion, housing an exchange, shall be completely equipped with auto- 
matic sprinklers. Buildings of fireproof construction shall be 
equipped from and including the lowest floor on which film is handled 
to the top of the building with an approved automatic sprinkler 
system. It is recommended that the sprinkler system extend through- 
out the building. 

(222) Exchanges shall be provided with one or more independent 
rooms to be used exclusively for receiving and delivering film, and 
also one or more separate rooms for the purpose of inspecting, ex- 
amining, and repairing film, and one or more rooms for the storage 
of posters or other combustible materials. 

(223) Shipping Room. One or more vaults or cabinets shall be 
provided in connection with the receiving and shipping room of 
exchanges into which all film shall be placed and kept except during 
such time as is necessary for checking, sorting, and shipping. All 
film outside the vaults and cabinets, except while actually being 
handled, shall be kept in I. C. C. containers. 

NOTE. With the enforcement of the above general principles of operation, the 
total quantity of film, including that which is in I. C. C. containers, in the re- 
ceiving and shipping room of any exchange should ordinarily not exceed 100 to 
150 standard rolls or 50 to 75 double rolls, with a maximum limit, which should 
never be exceeded, of 300 standard rolls or 150 double rolls or 1500 pounds. 


(224} Quantity of Film. In inspection, projection, rewinding, 
and other rooms (not including shipping room) there shall not be in 
excess of 16 standard rolls or 8 double rolls for each person handling 
film in such rooms, of which not in excess of 2 standard rolls or one 
double roll for each person shall be exposed outside of closed con- 
tainers. All film in excess of this quantity shall be kept in cabinets 
or vaults. 

Section 23. Motion Picture Film Laboratories 
(See Part I, General Provisions, which also apply.) 

(231) All buildings used for, or housing a motion picture film lab- 
oratory shall be equipped throughout with automatic sprinklers. 

(232) Quantity of Film. In all the various work rooms in which 
film is handled (not including shipping rooms) the quantity of film 
not in containers shall not exceed two standard rolls per person han- 
dling film ; this should not be construed, however, as restricting the 
quantity of film which may be in process on printing, developing, or 
drying machines to two standard rolls. There shall not be more than 
10 standard rolls of film not in approved cabinets for each person 
working in such rooms; provided, however, that in developing rooms 
there shall not be more than 20 standard rolls of film not in approved 
cabinets for each developing unit. All film in excess of the above 
quantities shall be kept in cabinets or vaults. 

(233) Printing. On all future installations, unless printing ma- 
chines are so spaced that the distance from the film on any machine to 
that on any other machine is at least 6 feet, they shall be separated 
from each other by incombustible partitions of 3 / 8 -inch hard asbestos 
board or its equivalent in heat insulation and durability, and extend- 
ing from the floor to at least 3 feet above the top of the film on the 
machines. If partitions carried to this height would extend higher 
than 2 feet below sprinkler deflectors, they shall be built up to the 
ceiling. If partitions are extended to the ceiling one sprinkler head 
shall be located in each of the sections thus formed. In any event, 
sprinklers shall be so arranged that not more than two machines are 
dependent upon the protection afforded by any one head. 

(234) Drying. Drying machines of the cabinet type shall be of 
metal and wired glass. Heating units shall be located outside the 
cabinet, and shall be provided with thermostatic control so that the 
temperature in the cabinet shall not exceed 120F. 


(235} Waxing. Waxing film shall be done in a separate room. 
Waxing processes which, require the waxed film to be left exposed 
to dry shall be in a room used for no other purpose and not over 5 
such machines shall be located in one room. Not over 25 standard 
rolls or 25,000 feet of film shall be exposed at one time. 

(236} Projectors. Not more than 5 motion picture projectors shall 
be located in one room, unless the projectors are of a type using 
incandescent electric lights of not over 25-watt size when not more 
than 10 projectors shall be located in one room. 

(237} Shipping Room. (a) The shipping room shall be separated 
from the rest of the building by partitions constructed in accordance 
with the provisions of sub-section 112. No other process than pack- 
ing of film shall be conducted in the shipping room. 

(6) Not over 500 standard rolls of film shall be in a shipping room 
at one time, of which the quantity not in shipping cases shall not 
exceed 250 standard rolls. See sub-section 181. 

Section 24. Motion Picture Studios 
(See Part I, General Provisions, which also apply.) 

(241} Buildings housing motion picture studios shall be completely 
equipped with automatic sprinklers, except that upon specific ap- 
proval of the inspection department having jurisdiction, sprinklers 
may be omitted in rooms of a construction having a fire retardant 
classification of not less than one hour and used only for housing valu- 
able electrical equipment and in which no film or other hazardous 
materials are handled or stored. 

NOTE. For sound recording studios and motion picture titling studios see 
Section 21. 

(242} (a) On the studio stage there shall be no film except in the 
magazines of cameras or sound recording apparatus of which there 
shall not exceed two magazine units for each camera or recording 

In other sections of a motion picture studio the quantity of film not 
in containers shall not exceed one standard roll per person handling- 
such film. 

(b} Extra, loaded magazines may be kept in special magazine 
room. The number of loaded magazines in one such room shall not 
exceed 50. They shall be kept on open metal racks. Such rooms 
shall be used for the loading and storing of magazines only. 


(c) In all sections of a motion picture studio other than the stage 
and magazine rooms, the quantity of film in a room shall not exceed 
10 standard rolls per person handling film, in addition to what may be 
kept in cabinets. 

(d) All film in excess of the quantities permitted above shall be 
kept in cabinets or vaults. 

(243) Vaults and/or cabinets shall be located in a section separated 
from the studio stage by partitions. (See sub-section 112.) 

(214) Sections of a studio in which the work is of the same general 
character as that in a laboratory shall be governed by the provisions 
of Section 23, Laboratories. 

(245) (a) Carpenter shops, property storage rooms, costume 
rooms, and dressing rooms shall be separated from the studio stage 
by fire partitions constructed of 8 inches of brick, or of some other 
construction of incombustible materials and suitable stability, having 
a fire retardant classification of not less than 2 hours as determined 
by the Standard Fire Test. 

(b) Only such openings as are necessary shall be provided in such 
fire partitions and shall be protected by approved fire doors of a type 
suitable for use in Class B situations as defined in the Regulations for 
the Protection of Openings in Walls and Partitions against Fire; open- 
ings of larger than standard size may be provided when necessary if 
protected by approved oversize fire doors. 

(246) Materials constituting a permanent finish or interior sur- 
facing on ceilings, permanent partitions, and walls, and used to reduce 
the reflection or transmission of sound, shall, if not incombustible, be 
or treated so as to be of an approved slow burning composition or 

(247) All fabrics of monk's cloth, canvas, muslin, burlap, silk, 
satin, velvet, velour, or similar material suspended from ceilings or 
gridirons or hanging against walls or partitions, and all backdrops, 
cycloramas, and other theatrical appurtenances constructed in whole 
or in part of muslin, canvas, burlap, and all artificial or natural trees, 
shrubbery, grass mats, straw, hay, and similar combustible material 
shall be painted, sprayed, or saturated with fire retarding or flame 
proofing material or otherwise rendered safe against fire; provided, 
however, that furnishings of silk, satin, velvet, and velour which are 
used in sets and included in the photographing of scenes need not be 
so treated. 


During the Conventions of the Society, symposiums on new motion picture appara- 
tus are held, in which various manufacturers of equipment describe and demonstrate 
their new products and developments. Some of this equipment is described in the 
following pages; the remainder will be published in subsequent issues of the Journal. 


Motion pictures which have been produced by exposing single frames at regu- 
larly spaced intervals and then projecting the film in the ordinary way constitute 
a valuable method of studying growths or movements which take place at a 
relatively slow rate. By way of contrast to our familiar slow-motion pictures, 
such movies may be called "fast-motion" pictures because they apparently speed 
up the action and cause, it to appear to take place faster than it actually did. 
These "time-lapse" pictures find their most frequent application in the study of 
growing plants or cells and similar subjects; for the growth or movement which 
required weeks or months may be condensed to a few moments on the screen 
and is available for repetition and study. 

The essentials for time-lapse photography are a motion picture camera that 
can be actuated one frame at a time and an automatic method of actuating such 
a camera at intervals which are adjustable to suit the rate of growth or motion 
being studied. As far as the camera is concerned, the Cine-Kodak Special is 
well suited for this type of work, and the apparatus to be described has been 
built to be used in conjunction with it. An automatic method of actuating the 
camera at desired intervals is a necessity, since in most cases it would be entirely 
impracticable for an operator to do this. The purpose of this paper is to describe 
briefly an apparatus which has been designed by the Eastman Kodak Company 
to meet such a need. The principal requirements for such an outfit are flexibility 
and convenience. 

In some cases it may be desirable to make pictures a few seconds or a fraction 
of a second apart, because the movement being studied is relatively fast. In 
other circumstances, the interval between successive pictures may have to be 
minutes or even many hours in order to produce the desired amount of action 
on the screen in a reasonable length of time when the film is finally projected as 

* Presented at the 1939 Meeting at New York, N. Y. ; received September 
30, 1939. 

** Eastman Kodak Co., Rochester, N. Y. 



a motion picture. In addition, such an outfit should be convenient to use and 
should be easily portable. Finally, since so many subjects will require artificial 
illumination, it is important to have an automatic method of turning the lights 
on for the exposure and turning them off afterward, no matter what the interval 
between pictures may be. The equipment which has been constructed to meet 
the above requirements is known as the Electric Time Lapse Outfit. It consists 

FIG. 1. The Electric Release Control for the Cine- 
Kodak Special, consisting of the Electric Release, which 
is mounted on the camera, and the Electric Release Con- 
trol box, which supplies the necessary impulses to control 
the camera action. 

of two parts, the Electric Release Control and the Interval Timer, which will be 
described separately. 

The Electric Release Control. The Electric Release Control is shown in Fig. 1. 
It consists of the Electric Release, which attaches to the Cine-Kodak Special, 
and the Electric Release Control box. The Electric Release contains an electro- 
magnet, to which electrical energy is supplied by the Electric Release Control 
box. It is mounted on the camera, as shown in the illustration, in such a way 
that when energized momentarily the electromagnet pulls an armature away 


from a butterfly cam placed on the one-frame shaft of the camera and allows the 
shaft to rotate either a half or a full turn under the camera's own power, according 
to the position of a control lever. A half-rotation of the shaft changes the camera 
shutter from closed to open, or the reverse. This action, which corresponds to 
that of an ordinary "still" camera shutter when set on Time, permits time ex- 
posures to be made. In another position of the control lever the shaft is allowed 
to make a complete revolution for each impulse supplied from the Electric Re- 
lease Control box. This results in instantaneous exposures. 

The Electric Release Control box houses the electrical equipment needed for 
the control of the electromagnet in the Electric Release. By means of proper 
settings it is possible to supply momentary impulses to the Electric Release in a 
variety of ways. Either instantaneous, time, or bulb exposures can be made 
manually by the operator, or instantaneous exposures can be made automatically 
at intervals of from l /^ to 6 seconds, through the use of a condenser-resistance 
timing circuit. Also, it is possible for either the operator or some external 
timing device to initiate single, automatically timed exposures which may be 
set throughout a range of */4 to 6 seconds. And, finally, a remote control of the 
ordinary camera action is provided by allowing the electromagnet to be energized 
continuously so that the camera is free to run in the normal way. 

Although power is normally supplied from four self-contained flashlight cells, 
provision is made on the panel for connecting an external 6-volt battery for long, 
unbroken runs. Also, a hand-control switch is provided which can be used in 
place of the control button on the panel, thus allowing the operator considerable 
freedom of movement while still retaining control of the camera. 

The Electric Release Control thus presents in itself complete manual and 
limited automatic control of the Cine-Kodak Special in applications where indi- 
vidual film frames are exposed at intervals or where remote control of the ordinary 
camera action is desired. It is especially intended for animation work, growth 
studies, and similar projects. Although it will automatically take instantaneous 
pictures spaced no more than six seconds apart, these intervals are quite satis- 
factory for many types of growth or other work. When used in conjunction 
with a timer and lamp control, such as will be described below, its range and 
usefulness are extended so that it will cover almost any conceivable application. 

The Interval Timer. The Interval Timer is an instrument which has been 
designed primarily for use in conjunction with the Electric Release Control and 
the Cine-Kodak Special. Its functions are to actuate the Electric Release 
Control at intervals, and thus cause either instantaneous or time exposures to 
be made and, in addition, to provide means for turning on and off the lights 
required in many cases. Like the Electric Release Control, the Interval Timer 
has self-contained flashlight cells, but facilities are provided for the use of an 
external battery when needed. As in the Electric Release Control, the basic 
timing circuit is achieved by means of a condenser being charged through a 
resistance. In the Interval Tinier the timing circuit provides regularly spaced 
impulses according to its setting, which come either */4 second or V minute 
apart. Through the use of a fixed multiplier which introduces a factor of 60, 
a basic interval of l /^ hour is also possible. The fundamental idea of the instru- 
ment is to multiply one of these basic times by a predetermined factor in order 
to achieve a different longer interval. The impulses are used in a variable multi- 

Mar., 1940] 



plier to index a ratchet wheel forward, one tooth at a time, by means of an ac- 
tuating relay. After a certain number of impulses, a contact on the wheel strikes 
an adjustable contact which causes the wheel to reset automatically to its zero 
position. This action of resetting determines the completion of the interval or 
cycle. It turns on the lamps and subsequently starts the exposure via the 
Electric Release Control, or in case the lamp control is not being used, initiates 
the exposure directly. The number of impulses necessary for resetting is deter- 

FIG. 2. The complete Electric Time Lapse Outfit, 
shown with the Cine-Kodak Special. The Interval 
Timer is in the foreground, the Electric Release Control 
box is at the left, and the Electric Release is mounted on 
the camera. 

mined by means of a dial knob on the panel, and may be set from 1 to 96. This 
setting, together with the selection by another knob of the basic time which is 
to be used, determines the interval for which the machine is set. At the end of 
every such interval the machine will reset automatically, initiate an exposure 
through the Electric Release Control, and proceed to time the next interval. 
The scales are so arranged that the instrument may be set to initiate exposures 
at intervals from V< second to 24 seconds in 1 /4-second steps, from Y minute to 
24 minutes in 1 /4-minute steps, and from l / hour to 24 hours in Vi-hour steps. 

For almost all time-lapse work which extends longer than a few hours, it is 
desirable to have the subject illuminated by artificial light so that the exposure 


and general lighting may be kept uniform. For pictures spaced farther apart 
than about half a minute or so, it is equally undesirable to have the lamps burning 
continuously, particularly if they are of the Photoflood type and considerable 
light is required. For this reason it was decided to incorporate into the Interval 
Timer an automatic lamp control so that the lamps would be turned on before 
the exposure is made, and turned off afterward, regardless of the interval between 
pictures. This is accomplished by providing a small synchronous line-voltage 
motor which controls the lamp circuit. When the end of the cycle is reached 
and the wheel resets, the lamps are turned on, and shortly thereafter the Electric 
Release Control is activated, thus causing the exposure to take place. Eight 
seconds after the lights go on they are automatically cut off, regardless of the 
duration of exposure. This is to cover the longest time exposure possible 
namely, 6 seconds. In case pictures are being made at intervals of a few seconds, 
the lamp control circuit is not used and the lights are simply plugged directly 
into the line. Under such conditions, or where the lamp control is not desired, 
the motor is not used, and therefore no power line to the Interval Timer is neces- 

The Complete Time-Lapse Outfit. Although the Electric Release Control may 
be used alone with the Cine-Kodak Special in many instances, it is the addition 
of the Interval Timer to form the complete Electric Time-Lapse Outfit which 
provides the most convenient way of meeting almost any growth study or similar 
problem. The complete outfit is illustrated, with the Special, in Fig. 2. Opera- 
tion is simple and there are relatively few steps involved in setting up the equip- 
ment to make a time-lapse film. The Electric Release, together with the cam 
which governs the action of the one-frame shaft, must be mounted on the camera 
and proper connections made to the Electric Release Control box, and from it 
to the Interval Timer. In case lamps are to be used, they are plugged into an 
outlet on the Interval Timer panel, and the power line which furnishes the lamp 
current is connected to a corresponding receptacle. Assuming the necessary 
camera and lighting adjustments to be made, the actual exposure time desired 
is set by means of knobs on the Electric Release Control box panel. This may 
be either instantaneous, varying with camera settings from Vioo second to Vzo 
second, or a time exposure which may be set throughout a range of V< second to 
6 seconds. The interval desired between the beginning of successive exposures 
is set by means of the two control knobs on the Interval Timer panel, as explained 
previously. It is important to note that the exposure settings and the interval 
time settings are entirely independent. All that remains is to turn a control 
switch, and the complete outfit will function without any attention whatsoever, 
as long as the spring motor in the camera is not run down. During the progress 
of the run any necessary changes in the adjustments to compensate for changes 
in growth rate or lighting may be easily made without interrupting the action of 
the outfit. Inasmuch as both units have self-contained batteries, the outfit is 
easily portable. Unless power for the lamps is wanted, a power line is not neces- 

The complete outfit as described thus provides all the equipment (not including 
camera) necessary for making time-lapse films. Although a great many uses of 
such an outfit lie in the field of growth study work, it has, of course, applications 
in many other fields as well. For example, in recording the progress of a con- 


struction project, in recording the readings of meters or controls at desired 
intervals, in traffic-control study, and in fact, in any application where the 
motion or growth occurs at such a rate that it is convenient to condense it for 
the purposes of study, such equipment should be extremely useful. 


MR. FLORY: What load will the interval-timer accommodate on the lamp 

MR. EATON : It is rated at 10 amperes, 1 15 volts. 

MR. FLORY : Can you use an external source of power, such as a 6-volt battery? 

MR. EATON: Either of these boxes can be used either with its own self-con- 
tained batteries, or with an external battery. The same 6-volt battery will func- 
tion for both of the units. 

MR. BRADY: Could the outfit be used for submarine work; say, in a diving 

MR. EATON: Certainly. The only limitation is the fact that the camera 
spring motor will run down after a certain length of time, depending upon the in- 
terval between the pictures. The camera has a long run about 40 feet of 16-mm 
film but if pictures are taken every second or thereabouts, the film will be used 
up very rapidly. On the other hand, if the interval is longer say, one picture an 
hour the camera will operate for a long time without attention. 

MR. HOLSLAG: Does the apparatus permit the use of a motor drive? 

MR. EATON: There is a motor drive, of course, as an accessory for the Cine- 
Kodak Special, but it would have to be adapted to be used with this time-lapse 
outfit. If you had a special problem that you wanted to work out, it could be 
done, using the outfit as a basis. 



The splicer to be described was developed for use on high-speed, sprocketless, 
35-mm developing machines. Above all else, such a splice must be dependable ; 
that is, it must not catch on guards or blow-offs, and must not pull apart in spite 
of passing around hundreds of small pulleys. Metallic fastening devices localize 
the stresses which are apt to start tears. By using a special adhesive tape to 
make the splice, the stress is distributed over the full width of the film. To 
prevent the tape from being soaked loose in the developing solutions, its "linen" 
base is waterproofed, and a hole is punched through the center of the film at 
each edge of the tape. The tape thus sticks to itself through these holes, pre- 
venting the edge of the tape from lifting when the emulsion swells. The ends 

* Presented at the 1939 Fall Meeting at New York, N. Y. ; received October 
15, 1939. 

** Eastman Kodak Co., Rochester, N. Y. 



of the film to be spliced are placed end-to-end with a space of Vi mcn between, 
and the tape is passed tightly around both sides, overlapped in the center of the 
film, and pressed down firmly. The splice will withstand a pull of 50 pounds, is 

FIG. 1. The first model. 

FIG. 2. The final splicer. 

smooth, and thus will not catch in the machine. Being quite thin, it will pass 
through blow-offs, and as many as forty splices have been taken up in a 2000-ft. 
roll without damaging the film or making the roll badly out of round. 


Two splicers have been designed to make this splice. The first one (Fig. 1) 
built prepares both films simultaneously in a punch and die, dispenses the tape 
and creases it sharply around the sides of the film, and has a device for squeezing 
the splice between two rubber pads with great force. This gives very intimate 
contact between the adhesive and the film and forces the tape to stick to itself 
through the holes in the center of the film and the perforation areas. The edges 
of the film are also cut away about Ve4 inch where the tape passes around the 
film so that the splice does not project even the thickness of the tape. The 
splicer is fitted with a small green safelight which projects a small spot of light 
on the splicing position and comes on automatically when the end of the film 
passes between two rollers. The same action causes a solenoid-operated brake 
to hold the film so that the end is in approximate position for splicing. An 
experienced operator requires 12 seconds to make a splice in the dark using this 

Although this first splicer was satisfactory from a performance standpoint, its 
cost was too great. To meet this objection, one of the authors designed the second 
model (Fig. 2). This model may be built at about one- tenth of the cost of the 
former. It works as follows: 

The end of the roll of film is placed in the punch and die, and three 5 /i<rinch 
holes spaced 1 / 2 inch apart are punched on its centerline and the end trimmed 
square. The two holes away from the end are then placed over two of the 
pegs at the front of the splicer. The trailing end of the leader is treated in a 
similar manner, the same perforators making the holes, but a knife on the other 
edge of the die is used to trim the end. The leader is then placed on the re- 
maining pair of pegs which line it up with the film so as to leave a Vie-inch space 
between the ends. A piece of tape about 3 1 /* inches long which had previously 
been placed symmetrically with respect to this space between the films, and 
sticky side up, is wrapped tightly around the film. The splice is then removed 
and the tape pressed firmly into contact with the film and with itself through the 
holes in the film. The space between the ends of the film helps to make the 
splice flexible for passing around pulleys. 


MR. CRABTREE: What are the advantages of this type as compared with other 
types of splices, as, for example, the Mercer patch, or staple, or actual cementing 
of the films? 

MR. BEGGS: The splice is absolutely safe in a developing machine. It will 
not pull apart, and in leader it will go through hundreds of times without damage. 

MR. PALMER : Have you tried to make a splice and maintain the sprocket hole 
so that the film will run over a sprocket? 

MR. BEGGS: We do not recommend the splice for sprocket machines, although 
we are actually using it on a low-speed sporcket machine. It seems to behave all 
right. All you have to do is to splice with the sprocket holes in register and not 
perforate the tape. It will climb over the sprocket all right but slips a tooth once 
in a while. 



The editors present for convenient reference a list of articles dealing with subjtcts 
cognate to motion picture engineering published in a number of selected journals. 
Photostatic copies may bt. obtained from the Library of Congress, Washington, D. C., 
or from the New York Public Library, New York, N. Y. Micro copies of articles 
in magazines that are available may be obtained from the Bibliofilm Service, Depart- 
ment of Agriculture, Washington, D. C. 

American Cinematographer 

21 (January, 1940), No. 1 
Camera Technique Dominates Filming Results (pp. 11, 

Amateur Progress in 1939 Exceeded Professional (pp. 

16-17), Pt. I 
Studying Photoelectric Exposure Metering (pp. 18-21, 

44-45), Pt. II 
DuPont's Superior II Combines Speed, Fine Grain, 

Wide Latitude (p. 22) 
Composition Is Simple Perhaps but Very Important 

(pp. 27-28, 39) 
Obtaining Increased Illumination from Fine Grain Film 

Recording (p. 36) 
Simple Changes Improve Camera Equipment (pp. 38, 


British Journal of Photography 

86 (November 24, 1939), No. 4151 
Progress in Color (pp. 695-696) 

The Projector in Detail, Sound-on-Film Projectors 
(pp. 696-698) 
Progress in Color (pp. 708-709) 

86 (December 8, 1939), No. 4153 
Progress in Color (pp. 723-724) 


12 (December, 1939), No. 12 
Disc-Cutting Problems (pp. 17-19) 

General Electric Review 

42 (December), No. 12 

Spectral Distribution of Radiation from Lamps of Vari- 
ous Types (pp. 540-543) 










International Photographer 

11 (December, 1939), No. 11 
Multiplane Camera for "Pinocchio" (p. 4) 
Television Camera Operation (p. 17) 


21 (November, 1939), No. 11 

Normung der Tonspur beim 16-Mm Schmalfilm (Stand- 
ardization of Sound-Track for 16-Mm Film) (pp. 

Die Verwertung von Altfilm-Material (Use of Old Film 
Stock) (pp. 250-253) 

Die Reichsverordnung Uber den Sicherheitsfilm vom 30. 
Oktober 1939 (Safety Film Ordinance of October 30, 
1939) (pp. 253-254) 

International Projectionist 

14 (November, 1939), No. 10 

The Effect of P. E. Cell Cable on Sound-Film Reproduc- 
tion Quality (pp. 7-8, 11) 

Assaying Projector Carbon Performance (pp. 12-15) 
Safekeeping the Picture Industry (pp. 15-16) 
Technicolor Adventures in Cinemaland (pp. 17-19, 23- 

Motion Picture Herald (Better Theaters Section) 
137 (December 9, 1939), No. 10 

Economical Mirror Maintenance for Effective "Suprex" 
Carbon Arcs (pp. 35-36, 38, 43) 

Photographische Industrie 

37 (November 15, 1939), No. 46 

Untersuchung uber ein Tontrennungsverfahren (In- 
vestigation on a Sound Separation Method) (pp. 1142- 
1143), Pt. I 

37 (November 22, 1939) , No. 47 

Untersuchung uber ein Tontrennungsverfahren (In- 
vestigation on a Sound Separation Method) (pp. 1156- 
1158), Pt. II 












Officers and Committees in Charge 

E. A. WELLIFORD, President 
S. K. WOLF, Past-President 
W. C. KUNZMANN, Convention Vice-President 
J. I. CRABTREE, Editorial Vice-President 
S. HARRIS, Chairman, Papers Committee 
J. HABER, Chairman, Publicity Committee 
H. GRIFFIN, Chairman, Convention Projection 
E. R. GEIB, Chairman, Membership Committee 
H. BLUMBERG, Chairman, Local Arrangements 

Reception and Local Arrangements 






Registration and Information 

W. C. KUNZMANN, Chairman 


Hotel and Transportation 

E. O. WILSCHKE, Chairman 





J. HABER, Chairman 

S. HARRIS P. A. McGuraB 


Convention Projection 

H. GRIFFIN, Chairman 





Officers and Members Projectionists Local 310, IATSE 


Banquet and Dance 

M. C. BATSEL, Chairman 




Ladies 1 Reception Committee 

MRS. O. F. NEU, Hostess 

assisted by 




Miss L. A. MOVER, Social Director, Chalfonte-Haddon Hall 


Headquarters. The headquarters of the Convention will be the Chalfonte- 
Haddon Hall, where excellent accommodations have been assured, and a recep- 
tion suite will be provided for the Ladies' Committee. 

Reservations. Early in March room reservation cards will be mailed to mem- 
bers of the Society. These cards should be returned as promptly as possible 
in order to be assured of satisfactory accommodations. 

Hotel Rates. Special rates have been guaranteed by the Chalfonte-Haddon 
Hall to SMPE delegates and their guests. These rates, European plan, will be 
as follows: 

Four Lower 

Floors Ocean View Ocean Front 

Room for one person $3.50 $4.00 $5.00 

Room for two persons 6.00 7.00 8.00 

Parlor Suite, for one 10.00 12.00 . 14.00 

Parlor Suite, for two 14.00 16.00 18.00 

(All bathrooms at Haddon Hall have hot and cold running fresh and salt 


If American plan rates are desired the hotel room clerk should be advised 
accordingly when registering. An additional charge of $3 per day per person 
will be added to the above-listed European rates for three daily meals, American 
plan. Members and guests registering at the hotel on the American plan will 
pay only $3 for the SMPE banquet scheduled at Haddon Hall on Wednesday 
evening, April 24th. If registered on the American plan, the clerk at registration 
headquarters should be advised accordingly when procuring your banquet 
tickets. , 

Parking. Parking accommodations will be available to those who motor to 
the Convention at the Chalfonte-Haddon Hall garage, at the rate of 5Q for day 
parking or $1.25 for twenty-four hours. These rates include pick-up and delivery 
of car. 

Registration. The registration and information headquarters will be located 
at the entrance of the Viking Room on the ballroom floor where the technical and 
business sessions will be held. All members and guests attending the Convention 


are expected to register and receive their badges and identification cards required 
for admission to all the sessions of the Convention, as well as to several motion 
picture theaters in the vicinity of the Hotel. 

Technical Sessions 

The technical sessions of the Convention will be held in the Viking Room of 
the Hotel. The Papers Committee plans to have a very attractive program on 
papers and presentations, the details of which will be published in a later issue 
of the JOURNAL. 

Luncheon and Banquet 

The usual informal get-together luncheon will be held in the Benjamin West 
Room of Haddon Hall on Monday, April 22nd, at 12:30 p.m. The forty-sixth 
Semi-Annual Banquet and Dance of the Society will occur on the evening of 
Wednesday, April 24th, in the Rutland Room of Haddon Hall an evening of 
dancing and entertainment for members and guests. 

Ladies' Program 

A specially attractive program for the ladies attending the Convention is being 
arranged by Mrs. O. F. Neu, Hostess, and the Ladies' Committee. A suite will 
be provided in the Hotel where the ladies will register and meet for the various 
events upon their program. Further details will be published in a succeeding 
issue of the JOURNAL. 

Entertainment ' 

At the time of registering, passes will be issued to the delegates of the Con- 
vention admitting them to the Apollo and Strand Theaters, by courtesy of 
Weilland and Lewis Theaters, Inc., and the Stanley and Virginia Theaters, 
courtesy of Warner Bros. Theaters. These theaters are in the vicinity of the 

Atlantic City's boardwalk along the beach offers a great variety of interests, 
including many attractive shops and places of entertainment. 


Convention Vice-President 



At a meeting of the Section held on February 14th at the Hotel Pennsylvania, 
New York, Messrs. P. C. Goldmark and J. N. Dyer of the Television Engineering 
Department of Columbia Broadcasting System presented a paper dealing with 
"Quality in Television Pictures." 

Present television standards specify certain factors that determine the appear- 
ance of a television picture only to a limited extent. Other factors, however, 
such as contrast, gradation, brilliance, and the shape of the scanning spot are 
fully as important and were discussed by the speakers. 

A photographic method of producing artificial pictures that permits varying 
several of these factors, was explained, and pictures were shown which were ob- 
tained by this method and which approach ideal quality within a given set of 

The next meeting of the Section will be held on March 14th at RCA Photo- 
phone Studios, 411 Fifth Avenue, New York. Mr. C. H. Cartwright of Massa- 
chusetts Institute of Technology will present a paper dealing with the new types 
of coatings applied to lenses to improve the optical qualities. The presentation 
will be accompanied by demonstrations. 


On January 26th, at a meeting held at the meeting rooms of the Western 
Society of Engineers, Chicago, the technical staff of the Bell & Howell Company 
presented a paper on the "Care and Maintenance of Motion Picture Lenses." 

On February 27th, Mr. W. C. Kalb of the National Carbon Company, Cleve- 
land, presented a paper entitled, "Projection Light Then and Now." 

Both meetings were well attended and interesting discussions followed the 


At the beginning of this year, the Society of Motion Picture Engineers became 
affiliated with the Inter-Society Color Council, in the capacity of a Member 
Society. The SMPE is represented on the Council by Messrs. G. F. Rackett, 
R. M. Evans, and F. T. Bowditch. Mr. Rackett is Chairman of the SMPE 
Color Committee. Mr. Evans is Chairman of the delegates to the ISCC. 

A meeting of the ISCC Executive Committee was held at New York on Febru- 
ary 20th, at which Mr. Evans was present. Reports of the activities of the ISCC 
will be rendered and will appear in the JOURNAL from time to time. Members 
of the Society who are interested in presenting problems dealing with color, or 
who would like to receive information from the ISCC on specific problems, are 
invited to communicate with Mr. Evans, care of the general office of the Society. 





At a recent meeting of the Admissions Committee at the General Office of 
the Society, the following applicants for membership were admitted to the Asso- 
ciate grade: 


145-30 Brinkerhoff Avenue, 

Jamaica, N. Y. 

401 N. Piedmont Street, 

Arlington, Va. 
Kodak, Ltd., 
The Works, 

Middlesex, England. 

2995 Marion Avenue, 

New York, N. Y. 
HOAR, H. G. 

RCA Photophone, Ltd., 
Electra House, 

Victoria Embankment, 
London, W. C. 2, England. 


Kodak Japan, Ltd., 
Central P. O. Box 247, 
Osaka, Japan. 

Aeneas Hall, 

University of Southern Calif., 
Los Angeles, Calif. 

Hawk-Eye Works, 

Eastman Kodak Company, 
Rochester, N. Y. 


651 W. Robinwood Street, 
Detroit, Mich. 


11148 85th Avenue, 

Alberta, Canada. 

O'DELL, H. J. 

203 Hawes Lane, 
W. Wickham, 

Kent, England. 
PELLY, E. P. L. 

Merton Park Studios, Ltd.. 
269 Kingston Road, 
Merton Park, 

London, S. W. 19, England. 

155 West 20th Street, 

New York, N. Y. 

Eastman Kodak Company, 

Rochester, N. Y. 
Cat Rock Road, 

Cos Cob, Conn. 
909 Avenue P, 

Brooklyn, N. Y. 


126 College Place, 

Ypsilanti, Mich 


160 Beverly Road 
Syracuse, N. Y. 




Volume XXXIV April, 1940 



The Adjustable Equalizer as a Tool for Selecting Best Response 
Characteristics E. S. SEELEY 351 

Solution Agitation by Means of Compressed Air 

C. E. IVES AND C. J. KUNZ 364 

Effect of Aeration on the Photographic Properties of Developers 


The Epoch of Progress in Film Fire Prevention . . A. F. SULZER 398 

New Motion Picture Apparatus 

Safeguarding Theater Sound Equipment with Modern Test 



W. S. STANKO 409 

Officers and Governors of the Society 424 

Committees of the Society 427 

Current Literature 433 

1940 Spring Convention at Atlantic City, N. J., April 22nd to 
25th 436 

Abstracts of Papers for the Atlantic City Convention 439 

Society Announcements 451 





Board of Editors 
J. I. CRABTREE, Chairman 




Subscription to non-members, $8.00 per annum ; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscription or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or Hotel Pennsylvania, New York, N. Y. 
Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 

West Coast Office, Suite 226, Equitable Bldg., Hollywood, Calif. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1940, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is not 
responsible for statements made by authors. 


* President: E. A. WELLIFORD, 30 East 42nd St., New York, N. Y. 

* Past-President: S. K. WOLF, RKO Building, New York, N. Y. 

* Executive Vice-President: N. LEVINSON, Burbank, Calif. 

** Engineering Vice-President: D. E. HYNDMAN, 350 Madison Ave., New York, 
N. Y. 

* Editorial Vice-President: J. I. CRABTREB, Kodak Park, Rochester, N. Y. 

** Financial Vice-President: A. S. DICKINSON, 28 W. 44th St., New York, N. Y. 

* Convention Vice-President: W. C. KUNZMANN, Box 6087, Cleveland, Ohio. 

* Secretary: J. FRANK, JR., 356 W. 44th St., New York, N. Y. 

* Treasurer: R. O. STROCK, 35-11 35th St., Astoria, Long Island, N. Y. 


* M. C. BATSBL, Front and Market Sts., Camden, N. J. 

* J. A. DUBRAY, 1801 Larchmont Ave., Chicago, 111. 
** A. N. GOLDSMITH, 580 Fifth Ave., New York, N. Y. 
** H. GRIFFIN, 90 Gold St., New York, N. Y. 

* P. J. LARSEN, 29 S. Munn Ave., East Orange, N. J. 

* L. L. RYDER, 5451 Marathon St., Hollywood, Calif. 

** A. C. HARDY, Massachusetts Institute of Technology, Cambridge, Mass. 

* H. G. TASKER, 14065 Valley Vista Blvd., Van Nuys, Calif. 

* Term expires December 31, 1940. 
** Trem expires December 31, 1941. 



Summary. That the problem of manipulating the response characteristics of 
sound reproducing systems in theaters has been significant is attested by the current 
practice of equipment companies of providing a variety of response adjustments as part 
of their standard equipments. However, prior to the establishment of this practice 
several thousand installations had been made. Continuation of the use of those sys- 
tems presents a service organization with the problem of determining as expeditiously 
as possible the optimum response characteristic in specific cases. 

The paper which follows describes the design of an adjustable equalizer system to 
permit selecting quickly, through dial switch adjustments at the listening point in the 
auditorium, any one of some twenty billion theoretically possible response character- 
istics. The paper describes the circuits employed and sets forth the reasons which 
underlie the choice of particular configurations. The accompanying illustrations 
show the several generalized types of component curves from which the overall insertion 
effect of any combination of settings may be predicted. For each of the types of sound 
systems most commonly encountered the method of insertion and the considerations 
governing the insertion of the adjustable equalizer are given. The paper concludes 
with a brief discussion of the types of problems whose solution has been facilitated 
through the availability and use of the adjustable equalizer. 

The necessity of manipulating the frequency response characteris- 
tic of theater reproducing systems arises in connection with two 
types of problems. The first type involves the determination of the 
best characteristic for a given type of horn system. The result of 
this determination forms the basis for the design of that system's 
standard corrective networks. The second type of problem involves 
correcting some aggravated deficiency of quality in a particular 
theater. Such theaters are essentially non-conformists, since practice 
that is successful in most theaters having the given type of system 
fails to produce satisfactory quality in the problem case. The result 
of the work in such instances is a special equalizer, provided the 

* Presented at a meeting of the Atlantic Coast Section at New York, N. Y., 
Dec. 13, 1938; revised manuscript received Aug. 30, 1939. 
** Altec Service Corp., New York, N. Y. 


352 E. S. SEELEY [J. s. M. P. E. 

trouble proves to be of such nature that a carefully selected frequency 
characteristic corrects it. 

The first type of problem warrants careful exploration of possibili- 
ties since care in selecting a standard curve is rewarded by successful 
application in many installations. Thorough study in this connec- 
tion implies, in addition to a proper variation of types of auditoriums 
and recordings, that a wide range of characteristic variability must be 
available ; the more readily the variations may be produced, the more 
confident the engineer will be that he has chosen the best solution. 

It is the obligation of a servicing organization to make every effort 
to obtain for the theaters serviced the best quality the reproducing 
equipment makes possible. In applying this underlying principle to 
equipments of early design, which embrace the majority of theater 
sound installations, the first type of problem has been encountered. 

Installations which present the second type of problem, that is, 
the non-conformists, may be divided quite conveniently by the writer 
of a paper into two classes : those which can, and those which can not 
be corrected by some variation from the standard response character- 
istic. Unfortunately, when confronted with a particular case an 
engineer often finds it a difficult matter to determine into which classi- 
fication the case falls. 

If it is assumed that the trouble is one which a variation in the 
response will correct, the usual procedure is to set up some charac- 
teristic as a starting point; formulate, from a sufficient amount of 
listening to selected test material, a judgment of the changes in char- 
acteristic which may effect the desired improvement; determine how 
these changes may be accomplished electrically by available adjust- 
ments or by the addition of new elements ; effect the circuit modifica- 
tion ; make the aural test again ; and then repeat the cycle over and 
over until a satisfying solution is obtained. The physical changes 
are made rather readily when they are confined to equalization steps 
built into the system, but when such adjustments prove inadequate 
or are not provided, the engineer attacks the amplifier with a soldering 
iron and a handful of resistors and condensers. He estimates what 
elements at some selected point in the amplifier will have about the 
right effect; makes the change; runs a response curve; finds he 
missed the mark ; changes some of the element values ; runs another 
curve, and then decides he is perhaps near enough to try another 
aural test. The difficulties that inhere, even under the most favor- 
able conditions, in securing the best result are aggravated by the 


length of the period between aural tests, the mounting cost of over- 
time, and fatigue, all of which greatly impair discrimination. 

The principal handicaps to the engineer working on problems of 
either type in the field may be stated in another way. They are the 
protracted time between aural tests required to make an electrical 
change, and failure to bring about the exact electrical change that is 
desired rather than merely something which more or less approximates 
it. The effectiveness of these two handicaps produced the desire for 
a highly adjustable equalizer. The equipment was visualized as a de- 
vice which could be inserted readily into any reproducing system, 
and provide controls located at the listening point in the auditorium 
for instantaneous production of predictable changes in the character- 
istic over a very wide range. Such an adjustable equalizer system 
has been built and has proved of great usefulness as an engineering 

In planning the response-varying function of the equalizer, it was 
decided that certain principles must be carried out if the equipment 
was to have maximum utility. A variety of curve types must be 
provided of such character and of sufficient diversity to permit, in 
combination, the production of an insertion curve of any reasonable 
shape without regard to the predicted requirements of any particular 
system. This principle precluded the use of available standard equal- 
izers built, with some latitude of adjustment, as components of exist- 
ing sound systems, and dictated the use of networks which would 
produce component curves suitable for building up overall curves of 
almost any type within a reasonable range, that is, exempting ex- 
treme amplitudes and extreme slopes. Such curves must be capable 
of variation in amplitude and location on the frequency axis. It was 
also decided that the adjustment should be in definite steps and of a 
kind that would permit the operator of the device to visualize the 
overall curve he is setting up or to set a curve already visualized. 

Nine adjustable equalizer sections were provided, of which eight 
give functionally different curves and one a duplicate. These curve 
types are given the following descriptive labels: Low-Frequency 
Rise, Low-Frequency Droop, Low-Frequency Cut-Off, Mid-Fre- 
quency Peak, Mid-Frequency Dip, High-Frequency Rise, High- 
Frequency Droop, High-Frequency Cut-Off No. 1, High-Frequency 
Cut-Off No. 2. 

The High-Frequency Rise and Low-Frequency Droop curves 
shown in Fig. 1 are similar in form, but the selector switches manipu- 



[J. S. M. P. E. 

6 7 8 

ft 7 10 FOR LC 


FIG. 1. Generalized relative response of equalizer sections which raise 
or depress one end of the spectrum. 

late them about at opposite ends of the spectrum. A curve of 
given amplitude and at a given location on the frequency axis is des- 
ignated on the selector switch dials in terms of the frequency at 
which the curve has half its ultimate amplitude, and the amount of 

April, 1940] 



FIG. 2. Generalized relative response of equalizer sections which raise 
or lower a limited portion of the spectrum. 

this half amplitude. Thus a particular curve may be labeled LD 4 
db @ 100, indicating a curve having an overall amplitude of 8 db and 
an amplitude of 4 db at 100 cps. The slope of the curve is greatest 
at the indicated frequency. The curves as shown are drawn on a 

356 E. S. SEELEY [J. s. M. P. E. 

"relative frequency" scale, with their mid-points at 1.0. To locate 
them on a frequency scale for interpretive purposes their abscissas 
may be multiplied by the selected center frequency. 

The High Droop and Low Rise curves also shown on Fig. 1 are the 
inverse of the preceding curves, and are designated as to amplitude 
and mid-point location in a similar manner. 

The High Cut-Off curves are labeled with the frequency of 3-db 
attenuation and attain an ultimate slope of 6 db per octave. Two 
such sections are independently available, and when both are used at 
the same setting the equalization afforded by the combination at the 
indicated frequency is 6 db and the ultimate slope is 12 db per octave. 
This type of cut-off was recognized as inadequate for some two-way 
horn systems, but such systems are uniformly provided with low-pass 
filters which may be cut in or out readily. 

Two Low Cut-Off settings of a 12-db-per-octave curve were in- 
corporated and are controlled by a switch devoted principally to 
Low Droop curves. 

The Mid-Frequency Peak curve types are shown in Fig. 2. These 
are typical symmetrical resonance curves and are obtained by a 
series-resonant four-element two-terminal network designed to vary 
the curve in amplitude, spread, and frequency of resonance. Ampli- 
tudes of 2, 4, and 6 db are available. Spread, which is defined quan- 
titatively as the ratio of the two frequencies at which the curve has 
hah its maximum value, may be set at 1 / t , 1, and 2 octaves. The 
curves may be centered at any one of eleven frequencies distributed 
at fairly equal intervals from 200 cps to 5000 cps. By connecting 
an external decade condenser box to a pair of terminals provided, the 
peak may be placed at any frequency from 150 to 8000 cps with little 
departure from the indicated amplitude and spread. For these curves 
as for all the others, the settings provided result in curves which, 
within small tolerances, conform with the curves shown in Fig. 2. 
The half-octave 6-db combination is not available due to limitations 
resulting from the Q of the coil used. 

The Mid-Frequency Dip is the inverse of the peak. This network 
is similar to the Peak network but is connected across rather than in 
series with the circuit. The curves given by this section are similar 
to the Mid-Frequency Peak curves, and similar combinations of am- 
plitude and spread are available plus the 6-db Yj-octave combina- 

The insertion characteristic of any section is almost completely 

April, 1940] 






^ O U ^^ ^^ 


5* * 

1C CD 00 O Ol 

1-H T-l 



5 * - 

^ tl J 2 

D 1 O O 
u o O 

8* 22 

1C O 1C 1C O 1C O O 

CO g ^ 

"2 CN '3 
a > 


rt -g cS 


^ - s 

- rt 


Q ^ ^ 


N 1 s 

." > 


1 ., i 

3 <n '*, 

o 1 O O 


s "i > 

j> o S 


1C O 1C 1C O O 1C 1C 

ft S - 
jj a-a 

*^ W CO 


a a ^ 


S 5 S 

5 i-H ,-H 


fa fa ^ 

M M CL, 

S S 2 

"O "O ^H 

c a 


i iiCiCOiCiCiCOO 

M JJ g 

te 2 01 01 


i-l T-H CN CM CO 1C 

cS rt .2 

OJ D *J 

Hi fa a 

fa fa " 

S S w 

d . 
1 ry O 1C 
PU S O 00 

I-H ic 1C O iC^ 1C 1C O O 
1-1 TH OJ Ol CO 1C 


o" ic O 

a fi 01 ic 

I-H Ol ^1* O T-I CN -^ 

1 1 1 1 



i 1 Ol i 1 i 





J Q 


* T^ TP 1C 1C 3 

u jj 01 ic 



i-H C^ ^^ CM ^* i"H 


1* 1 ( 



358 E. S. SEELEY [J. s. M. P. E. 

independent of the settings of adjacent sections and, as a result, the 
sections may be combined and the result visualized without running 
a response curve. Even when the curves from several sections over- 
lap over an appreciable frequency range, the combined curve may be 
visualized readily, although examination of the generalized curves is 
necessary to permit accurate prediction. 

The switch settings listed in Table I represent the various curves 
available. While the adjustments permit 20 billion combinations, 
obviously, many of these probably would never be found useful. 
Although the adjustment range was considered adequate for all or- 
dinary requirements, an adjustment is occasionally desired that is 
outside the range provided. To accommodate such unusual require- 
ments, terminal posts have been included in three of the sections for 
external connection of adjustable resistance, capacity, or both. We 
provide for this purpose a decade condenser box having a range from 
0.0001 /if to 10 juf and an indicating tapered rheostat of broad range. 

Loss in volume level results when some of the equalizer settings are 
switched in and may be compensated by the level control. In gen- 
eral, High Rise and Low Rise cause a mid-range loss equal to the over- 
all amplitude of the curve although this generalization is subject to 
some qualification if the curve extends into the range that determines 
volume level of speech. High and Low Droop and the cut-offs 
have no important effect on volume level unless the curve extends to 
the voice-volume region. The MF Peak set outside the voice-volume 
region produces a loss somewhat greater than the amplitude of the 
curve, depending upon the selected combination of amplitude and 
spread. The MF Dip produces only a small loss unless it is located 
in the volume-determining range. 

The matter of phase distortion was considered early in the design 
of the equalizer system. The use of equalizers which introduce no 
phase delay was impracticable. Consideration was given to including 
several adjustable phase-delay sections for experimental purposes 
but space was not available within the dimensions to which we wished 
to confine the control unit. In general, the equalizer networks em- 
ployed in this system are of the types which would be used to equalize 
a system permanently to the curve found desirable by the adjustable 
equalizer, and therefore one may expect the equalization permanently 
installed to produce the same results, including any changes in phase 
characteristic, as were obtained with the variable equalizer. 

The equipment consists of a booth unit, a control unit located in the 

April, 1940] 





"1 "S 


t <=! 

t : 




' '"""""' 


r I 


t t= 
i t >o 







i 7 ^"" 1 

r 1 












. o 







MV1CIC loar. 


1 " $ 





3 rv 


360 E. S. SEELEY [J. S. M. P. E. 

auditorium, and interconnecting cables. The booth unit consists of 
an amplifier and power unit. The control unit contains the equalizer 
sections, an amplifier, volume control, output transformer, and out- 
put impedance selector switch. Three cables 100 feet long are re- 
quired: two for speech, and one for power services for the control 
unit amplifier. The entire schematic arrangement is shown in Fig. 3. 

The total amplification is divided into two parts. The first is 
placed ahead of the line to the auditorium in order that the signal 
transmitted may be of relatively high level to minimize danger of 
noise pick-up in this run. After a considerable loss is sustained 
from a number of equalizer sections, the second amplifier raises the 
signal level again for further equalization and the return to the booth 
and the system amplifier. 

The booth unit amplifier has three stages, resistance coupled, to 
deliver a maximum level of db relative to 0.006 watt. Feedback is 
employed to reduce harmonics generated in the second two stages 
and to reduce the output terminal impedance in order to minimize the 
effect of the capacity of the long shielded cable. A power unit having 
a high degree of filtering is included to permit a-c operation. The 
control unit amplifier consists of two stages embraced by a feedback 

For the equalizer sections design selection was available between 
two-terminal and four-terminal networks. With four-terminal con- 
stant-resistance networks, the characteristic of one section is inde- 
pendent of the setting of the adjacent sections without the necessity 
of isolation by intervening loss sections. Such networks, however, 
require double the number of elements, coils must be used in each 
section, and the result is a much larger, heavier, and more costly 
assembly. Further difficulties result from the necessity of procuring 
a large variety of coils having correct inductance and Q over the re- 
quired range of frequencies. For these reasons two-terminal net- 
works were chosen. The requirement of increased amplification to 
compensate for the loss in the isolating pads required between such 
sections was not regarded as a serious objection, since an amplifier is 
required in any event to compensate for the loss in either type of 

The point at which the equalizer system is inserted in the theater 
system varies with the type of system. In those systems which con- 
tain a low-impedance, low-level circuit, the equalizer is inserted at 
such a point, usually following the change-over device. Thus in most 


Western Electric and Motiograph Mirrophonic systems it is inserted 
between the fader and the main amplifier. In Simplex Four-Star 
systems, it is inserted just ahead of the main amplifier. In RCA 
Photophone systems it may be inserted in the low-impedance circuit 
if there is one, between sound-heads and main amplifier, and when this 
circuit is balanced to ground, it is temporarily changed to an un- 
balanced circuit. Photophone systems not having the low-impedance 
linkage are easily accommodated by removing the grid-cap connec- 
tion from the first tube and inserting the variable equalizer between 
the grid lead and the grid of the tube. Experience with a large variety 
of sound systems has revealed none to which the equalizer can not be 
readily adapted. 

Two limiting conditions must be kept in mind : 

(1) Input voltage levels exceeding 50 db relative to 1.73 volts 
require the use of the low-gain position of the input switch to avoid 
overloading the first amplifier. 

(2) The output hum level of the equalizer system is 100 db and 
in systems which do not have preliminary amplifiers the noise level 
begins to be disturbing if the amplifier has a gain of the order of 90 db. 

Considerable field experience has been accumulated through the 
use of this equalizer. The main handicap confronting an engineer in 
his attempt to obtain the most suitable response characteristic, 
namely, the difficulty inherent in achieving desired changes, is elimi- 
nated. The time, effort, and ingenuity otherwise so expended are 
profitably employed in attacking other phases of the problem. The 
significance attaching to the presence in the past of this manipula- 
tion handicap is emphasized by a brief review of instances on record. 
The record shows that in extreme cases, thoroughly experienced and 
qualified engineers spent from one to three weeks in a single theater 
before their conviction was established that the quality imperfection 
whose cause they sought was of such nature as not to be subject to 
correction by response characteristic manipulation. There have been 
cases, also costly to both the exhibitor and installation organizations, 
where a quality deficiency was assigned to causes other than response 
characteristic. Stage equipment was replaced, relocated, draped, or 
redraped to no avail. Subsequent trial showed the solution to reside 
in proper selection of the response characteristic. 

Certain types of distortion give rise to a conviction on the part of 
the engineer that correction must necessarily follow upon his making 
appropriate changes in response. Some problem houses can not be 

362 E. S. SEELEY [J. S. M. P. E. 

so compensated. The application of the adjustable equalizer to these 
problems quickly and convincingly proves the point. A pertinent 
illustration involved a house obviously wholly deficient in bass re- 
sponse although identical equipment in many other theaters provided 
adequate bass easily made excessive through additional low-end 
equalization. It quickly was found in the house under discussion that 
no practicable amount of low-frequency reinforcement of the charac- 
teristic produced a noticeable effect upon the acoustic response below 
100 cps. It was necessary to seek a "best" compromise between some 
apparent music bass and "heavy" speech reproduction through rein- 
forcing the range between 100 and 300 cps. The difficulty was as- 
signed to the light wall-construction employed. 

There have been instances in which correction of a quality defi- 
ciency was effected through response changes hi a part of the char- 
acteristic not usually associated with the observed deficiency. Such 
correction was evolved only by virtue of the ease with which changes 
of almost any conceivable kind could be made and the rapidity with 
which unsuccessful changes were rejected and others substituted. 
Such was found to be the case in a theater suffering from lack of 
"presence" where treatment in the region from 2000 to 3500 cps 
failed to produce the usual result, and the facility with which changes 
could be made led to elimination of the defect by reinforcing the 400 
to 600 cps range. 

An old lesson was re-learned forcibly at the outset of our program 
of study of and with the variable equalizer. That lesson was that 
efforts to improve the characteristic of a reproducing system hold 
promise of success only after the flutter content of the film propulsion 
mechanism has been reduced, where the chief component is 96-cycle 
modulation, to a value of the order of 0.4 or below. The effect 
produced by such flutter frequently is misinterpreted as resonance 
peaks in the 1500 to 3000 cps range. The TA-7421 flutter bridge, 
with which our field forces were equipped in 1936, affords a ready 
means for determining the flutter performance. Consequently, de- 
termination of the percentage of flutter present always precedes 
our use of the adjustable equalizer. 

The equalizer already has been used on sound systems of a variety 
of makes and types in many theaters. It has been used also in con- 
nection with the setting up of a binaural public address system, a 
synthetic sound system, and a high-quality disk reproducing system 
set up in a theater for a special demonstration, under the direct super- 


vision of the conductor, of recordings of an outstanding symphony 
orchestra. It is interesting to note that in the last case mentioned 
the equalization set up by the sound engineers was changed substan- 
tially by the conductor chiefly to tone down the strings and to build 
up the bass instruments to a dominance admittedly greater than that 
in the original playing of the orchestra. 

The listing of applications already found for the equalizer outside 
the specific purpose for which it was designed is evidence of the equal- 
izer's versatility and assures its permanence as a useful tool for 
sound technicians. 


Summary. Agitation of the developing bath by means of compressed air or 
other gas is most effective when the upward stream of air bubbles is made to follow 
the surface of the film. When the air is released at the bottom of the tank, the in- 
duced stream usually passes up through one portion of the tank without reaching all 
parts of the film. The stream can be made to follow the surface of the film in a verti- 
cal rack type developing machine by the use of an air distributing grid-work which 
discharges the air at a large number of points distributed along the film strands from 
bottom to top of the rack. 

In the course of photographic development, that portion of the de- 
veloping bath which is absorbed by the emulsion becomes altered by 
the depletion of its active ingredients and the accumulation of reac- 
tion products. This partially exhausted developer within the emul- 
sion has a lower reaction rate and must be replaced constantly by 
diffusion from the relatively unaffected developing solution adjacent 
to the emulsion surface. This solution, in turn, is renewed by means 
of developer brought to this region from the main body of the liquid 
in the containing vessel. The rate of renewal of developer within the 
emulsion by diffusion may be limited by the unavailability of fresh 
solution at the emulsion surface 1 but such a limitation would not be 
harmful if its effect were uniform over the whole emulsion surface. 
Unfortunately, this is not the case and some very undesirable effects 
occur unless steps are taken to prevent them. In general, the un- 
evenness of development of different areas results in differing densi- 
ties at points which have received equal exposures. In a large area 
which has had a uniform exposure a mottled or streaky appearance 
is evident. Exhausted developer accumulating near a high density 
area of considerable size may stream out over an adjoining area caus- 
ing a diminution in density. 

When the developer in the neighborhood of the film receives little 
agitation other than that resulting from the unidirectional motion of 

* Presented at the 1938 Spring Meeting at Washington, D. C. ; revised manu- 
script received March 18, 1940. 

** Kodak Research Laboratories, Rochester, New York. 


the film, well-defined streaks may be seen trailing away from high- 
density areas in a direction opposite that of the film movement during 
processing. The manner in which this condition affects sound rec- 
ords and sensitometric exposures arranged in the usual stepwise 
fashion has been studied by Crabtree and later by Crabtree and 
Waddell. 2 In this connection they used the term "directional ef- 
fects." The present work is concerned with improvement in de- 
velopment uniformity beyond the point where serious directional 
effects are shown in sensitometer tests. 

Several methods of promoting the renewal of the solution at the 
emulsion surface are possible, namely, (a) passing the film through the 
machine at high speed, (6) circulation of the developer by means of 
pumps and impinging the liquid against the film surface by means of 
jets, and (c) bubbling gases through the solution. No attempt is 
made in this paper to compare the effectiveness of the various meth- 
ods, the following data being restricted to the gas method of agitation. 

Stirring by Means of Gas Bubbles. The use of compressed air or 
other gases for the agitation of liquids ordinarily consists in releasing 
the gas bubbles in a stream at one or more points at the bottom of a 
vessel, from which point they rise through the body of the liquid and 
are released into the air at the surface. This method has had con- 
siderable use in chemical processes and has been used to some extent 
in motion picture work in the rack-and-tank process. Stirring by 
this means was particularly helpful in the Gaumont type continuous 
developing machine where the solution was contained in narrow ver- 
tical tubes. In these tubes, which were from 2 1 /* to 5 inches in di- 
ameter and as long as 6 feet, the stream of gas bubbles was, of neces- 
sity, constrained to follow a path where it agitated the liquid in con- 
tact with the film. Agitation by gas streams has apparently had 
little application to the present-day vertical rack developing machine. 
One reason for this may be the difficulty in controlling the direction 
of the stream in a tank of large cross-section. 

In order to avoid oxidation of the developer, nitrogen gas has been 
used to a limited extent but its use is expensive and, in many cases, 
unnecessary. With most developers the chemical effect of air used 
for agitation is hardly severe enough to raise any serious objection on 
the basis either of undesirable photographic effects or increased re- 
plenishing cost. 3 In the present work only compressed air was used. 

Air Agitation in a Vertical Rack Developing Machine. In order to 
study the action of air agitation in a developing machine, a special 

366 C. E. IVES AND C. J. KUNZ {]. S. M. P. E. 

tank was constructed of transparent Eastman acetate sheet by the 
method described by Hickman and Hyndman. 4 The support re- 
quired to keep the tank in shape when full was furnished by an open- 
work wooden frame as shown in Fig. 1. This tank, representing one 

FIG. 1. Transparent tank illuminated from the rear. 
Racks and vertical grid distributors in position. 

unit in a 30 to 40-feet per minute vertical rack type continuous proc- 
essing machine, was intended to accommodate two racks. It had 
inside dimensions of 5 J /4 by I3 l /t by 35*/2 inches deep. The tank was 
filled with water. Racks, cooling coils, etc,, were introduced from 
time to time as required. 


Method of Observation. It was found convenient to view the move- 
ment of airbells in the tank with the aid of illumination from one of 
the large sides. Viewed from the opposite side, bubbles in the liquid 
appeared somewhat darker than the bright surrounding field. Under 
these conditions, air bubbles released singly from the bottom of the 
tank were clearly visible in their course to the top surface of the 
liquid. Motion pictures made at rates up to 64 frames per second 
with this same illumination were found helpful in studying detailed 
patterns of movement of the airbells and for measurement of the rate 
of streaming. In order to determine the rate of bulk movement of 
the liquid in the tank carried along by the airbell stream, observations 
were made by introducing potassium permanganate solution into the 
tank at a suitable point. 

Movement of Air Bubbles of Various Sizes. It was found that air 
bubbles of varied size would be discharged through orifices in a rubber 
tubing made by cutting it with a point or chisel edge. At a low rate 
of flow, the large size bubbles 0/4 inch in diameter) were absent. 
However, smaller bubbles were present even at comparatively high 
rates (I 1 / \ cubic feet per minute discharging from 20 slits l /^ inch in 
length). By the use of glass or metal tubes with orifices of fixed size, 
the smaller bubbles (Vie inch and smaller) were largely eliminated and 
the range of sizes was rather narrow at any one rate of air discharge. 
The size of the bubbles could be increased somewhat by increasing 
the rate of flow and the size of the orifices. 

By discharging bubbles into the water at the bottom of the tank, 
singly or a few at a time, the dependence of the rate of bubble move- 
ment upon bubble size was observed Starting with bubble sizes 
which were almost microscopic, the rate of movement to the surface 
was very slow. When a strong flow of bubbles comprising a mixture 
of sizes, including the very small, had been passing through the tank 
for several minutes and was then stopped, the larger ones would dis- 
appear in a matter of a few seconds while the cloudiness resulting from 
the presence of numerous small bubbles would not be cleared away 
until several minutes had passed. The very small bubbles contrib- 
uted negligibly to the stirring of the developer and, therefore, were 
undesirable. As the bubble size was increased, the rate of rise of 
single bubbles through still water was also increased until a velocity 
of about 1 foot per second was attained. This velocity was observed 
with bubbles of approximately l /< mcn m diameter. Larger sizes 
had only slightly greater velocity. The individual bubbles larger 

368 C. E. IVES AND C. J. KUNZ [J. S. M. p. E. 

than about Vis inch in diameter rose with a quick, staggering motion. 

The rate and manner of movement of gas bubbles through a liquid 
has been the subject of a number of investigations. Many of them 
have had to do with the case of an elongated bubble having a diameter 
almost equal to that of a narrow tubular vessel in which the rate of 
movement was observed. In other cases of more immediate interest 
where relatively larger vessels were involved, velocities up to 1.3 feet 
per second were observed. 6 Some of the investigators found that 
the bubbles flattened and rocked and traversed helical paths depend- 
ing on their size and the diameter of the tube. In one case vortex 
rings were detected in the wake of the larger bubbles. 

Unquestionably the liquid near the bubble is moved about greatly. 
The significance of this is that a generally turbulent condition, which 
should be very helpful in obtaining uniformity of developer action, 
prevails throughout the path of a stream of bubbles. 

At the top surface of the liquid both vertical and horizontal move- 
ments of the liquid are readily observable. 

Channel Formation. When the rate of discharge of bubbles was 
increased, a general streaming of the liquid along the path taken by 
the bubbles was set up. Water propelled to the upper part of the 
tank along this path passed downward through another part of the 
tank and eventually reentered the ascending stream. As this action 
continued it was found that even if the airbells were released at nu- 
merous points along the bottom of the tank, they tended to run to- 
gether and form a relatively narrow column or channel. In the tank 
described, stream velocities within the channel reached a value of 
about 2 feet per second. At the same time, the movement of liquid 
downward and, to a certain extent, horizontally, in the remainder of 
the tank, diminished to practically zero in certain places. 

Factors Affecting the Shape and Position of the Channel. For an ex- 
tremely low rate of flow of air, where the bubbles rose singly or, at the 
most, a few at a tune, no distinctly recognizable channel existed. 
The channel formed, however, as soon as a substantial rate of air flow 
was reached (Vio cubic foot per minute per square foot of total hori- 
zontal cross-sectional area of the tank). Since the effect upon de- 
velopment uniformity produced by any less than this flow of air was 
quite poor, any improvement in air agitation would have to take the 
channeling tendency into account. As the air flow was increased 
still further, the channel became more clearly defined (Fig. 2), but 
the distribution was not improved with the air flow increased to as 

April, 1940] 




much as 3 cubic feet per second per square foot of horizontal cross-sec- 
tion of the tank. The top surface of the liquid was in such a state 
of strong agitation as to lead to the supposition that the whole con- 
tents of the tank were being stirred effectively which was not ac- 
tually the case as was found by viewing through the tank wall. 

Prior to the time of the present work, various arrangements of dis- 
tributing pipes and plenum chambers made to fit the bottom of the 
tank had been devised to improve the distribution of the air stream 
throughout the tank. The principal result of these efforts was to 
find a type of construction which would maintain equal air streams at 
the different orifices. This uniform- 
ity was best attained by the use 
of small orifices in rigid corrosion- 
resisting materials and with air 
pressures in the distributing pipes 
substantially greater than the ex- 
ternal pressure at the point of dis- 

Tests with Racks in Position. 
With the developing rack and the 
cooling coils occupying their nor- 
mal positions in the tank, observa- 
tions were made of the streaming 
patterns obtained with an air dis- 
tributing system consisting of a single 
metal pipe in which were drilled 
a number of small orifices. This 
was located on the bottom of the 
tank at the mid-line parallel to the 
longer dimension. With this dis- 
tributor operating in the normal way, the tendency to channeling 
was much the same as had been observed in an open tank although the 
presence of the racks caused a definite change in the position of the 
stream. Not more than 20 per cent of the film in the tank was sub- 
ject to the action of the rapidly moving stream. It was evident also 
that the liquid returning to the bottom of the tank was moving along 
some of the film strands while a condition approaching stagnation 
existed near others. The presence of the return path in this portion 
of the tank was a factor tending to cause the deflection of the much 
more rapid upstream to one end of the tank. 

FIG. 2. Path of bubbles rising 
from a horizontal distributor on 
the bottom of the tank. 

370 C. E. IVES AND C. J. KUNZ [J. S. M. p. E. 

This condition was improved somewhat by providing an additional 
return path in a portion of the tank beyond the ends of the racks. 
In order to make a passage into this space, large rectangular apertures 
were cut into the rack frame at a point near the upper surface of the 
liquid. Openings were also made in the bottom frame member of 
the rack so that the liquid after passing downward through the space 
at the end of the tank and thence along the bottom of the tank could 
rise vertically through the bottom of the rack. These modifications 
had the effect of moving the high-velocity channel toward the center 
of the tank where it came in contact with more film strands. 

Effect of Changes in the Distributors. An effort was made at this 
point to divide the stream by the provision of three separate distribut- 
ing tubes, one at the center and one at each side of the tank bottom. 
Better distribution was obtained at the bottom of the racks but the 
streams tended to run together toward the top forming the familiar 
channel. The results were no better when the distributing pipes 
were placed along the sides of the rack near the lower rack rollers. 
The location of the orifices in such a way as to deliver the air upward, 
downward, or horizontally toward the film was found to have little 
influence in directing the stream. The bubbles were simply released 
at the orifices and were not projected outward appreciably with the 
rates of flow employed. While it would have been possible by the 
use of baffles to prevent the union of the streams originating on each 
side of the rack, it was preferred not to modify the rack in this 

Porous Carbon Distributors. Another form of ah- distributor which 
has been used in electroplating apparatus is of the porous carbon 
type, 6 and consists of small particles of carbon held together with a 
carbon bonding material so that the resulting porous cake consists 
of 99 per cent carbon. 

Distributing units of various degrees of porosity designed to fit into 
the bottom of the test tank were furnished by the National Carbon 
Company for trial. Although the size of bubbles and the distribu- 
tion over the porous surface were quite satisfactory, the use of this 
type of distributor furnished no control over the tendency to channel 
in the upper part of the tank. 

A Means of Positioning the Channel. Since the channel, formed 
spontaneously, continues its existence by reason of the relatively low 
average density of the mixture of air and water thus formed, it was 
believed that the shape and position of the channel might be prede- 

April, 1940] 



termined if a region of reduced density could be established in the 
required position. 

Neglecting that portion of the film which lies in contact with the 
upper and lower rack rollers, the film in the developing machine lies 
along a vertical plane on each side of each rack. The channel 
through which the liquid should pass upward through the tanks 
would be contiguous to this plane. Thus, the cross-section of the 
channel might have the dimensions of 1 inch by 6 inches and, in 


FIG. 3. Details of vertical grid distributor: (a) distributor, (b) position 
of distributor relative to rack, (c) path of air through distributor, (d) 
method of introducing additional resistance to flow of aif in oversize distribu- 
tor tubes, (e) orifice in distributor tube, (/) construction details of distribu- 
tor manifold. 

length (vertically), should be equal to the length of the film strands 
on the side of the rack. 

It was found that such a region of reduced density could be formed 
by discharging small streams of air at a large number of points 
throughout this region from bottom to top of the rack. Air was de- 
livered to these points by means of a gridwork of small tubes erected 
near the side of the rack and at a distance of about V 4 inch from the 
film surface (Figs. 1 and 36). The individual tubes were arranged 
vertically (Figs. 3, a and 6). 

372 C. E. IVES AND C. J. KUNZ [J. S. M. p. E. 

The position of the streams or channels was established satisfac- 
torily by this means. The distributing tubes were located on ap- 
proximately vertical lines so that the film moving by at a small angle, 
passed a succession of the orifices displaced in small steps from one 
edge to the other across its width. The general direction of the 
stream was, of course, at the same small angle to the direction of the 
film movement but the movement of the developer at any one point 
was varied in direction because of the turbulent condition of the 

Development Uniformity. Photographic tests were made in the 
course of these changes by developing motion picture film which had 
been given a uniform flash. Development was to a degree consid- 
erably less than gamma infinity so as to produce a more easily detect- 
able degree of non-uniformity in the lack of adequate agitation. 
Improvement in uniformity was noted in successive alterations of the 
circulatory system as described previously. The greatest improve- 
ment in a single step occurred when the air agitation was first intro- 
duced. The successive modifications, while effecting a necessary de- 
gree of improvement, did not have as great an effect. A flow of 1 
cubic foot of air (atmospheric pressure) per minute at each side of a 
rack was found suitable with the grid finally adopted. Under actual 
working conditions with emulsions developed to their normal gammas, 
the density variations were small in all cases where agitation was em- 
ployed. It was attempted to make a graphical representation of 
density variation by means of a recording densitometer but this type 
of record was unsatisfactory, apparently because it showed only a 
profile of density variation along the line of scanning and, therefore, 
failed to show the whole pattern. Consequently it was necessary to 
judge results by projection and by viewing over an illuminator. In 
order to have sufficiently large density variations for judgment by 
projection it was necessary to "magnify" the variations in the original 
test sample by making a contrasty print. 

In practical processing work of an exacting nature, a degree of uni- 
formity satisfying current standards was obtained. The grid type 
distributor found helpful in the present case would be useful only in 
a machine in which the film path was as described above. Best re- 
sults have been attained when development was followed by the use 
of an acetic acid stop bath. 

Construction of the Air-Distributing Grid. In a grid work arranged 
as shown in Fig. 3, a and b, the air is discharged from the orifices lo- 


cated at 3-inch intervals along the length of the tubes marked T. 
Air is supplied to these tubes from the manifold pipe M at the bottom 
of the grid which, in turn, is connected to the supply tube S (Fig. 3a). 
The tubes receive their physical support from the manifold at the 
bottom and the horizontal member at the top. These are supported 
in turn by attachment to the supply tube S and a tubular member on 
the opposite side. 

Air is discharged at any particular point in the tank only after the 
pressure at the inside of the tube exceeds the resisting force outside 
the aperture which includes the hydrostatic head. Since the head is 
dependent upon the depth of the liquid, the greatest air pressure is 
required at the orifices located near the bottom of the tank. For this 
reason, the air was brought through the supply tube to the bottom 
manifold and then passed upward as required through the tubes (Fig. 
3c). With a total flow of air of about 1 cubic foot per minute (at- 
mospheric pressure) from each gridwork, a supply tube of Vs-mch 
inside diameter ( 3 /ie-inch outside diameter) was found ample. To 
facilitate welding at the several points on the manifold, this was made 
of 3 /ie-inch inside diameter material. 

Distributing grids were constructed with Vie-inch inside diameter 
tubing T which, by reason of its small size, furnished sufficient dis- 
tributed resistance to produce the necessary pressure gradient along 
the length required to compensate for the varying hydrostatic head. 
At the rate of discharge employed, the pressure gradient in distribut- 
ing tubes of larger diameter was not sufficient to compensate for the 
rate of change in hydraulic pressure from bottom to top of the tank. 
As a consequence, the bulk of the air was discharged at the top of the 
tube, the lower orifices being ineffective. 

If only a larger size of tubing is available, the required additional 
resistance can be introduced by constricting the tubes as shown in 
Fig. 3d, thus creating several zones of successively diminishing pres- 

The orifice holes are 0.030 inch in diameter and are made by hand 
by the use of a hard steel punch. Before the hole is punched the tube 
wall is thinned down to Vw inch by means of a Ys-mch drill (Fig. 3e). 

Stainless steel (18 per cent chromium-8 per cent nickel) tubing 
in the required sizes is available on special order from the regular 
sources of supply. The most desirable type of construction consists 
of welding wherever possible. With the smallest distributing tubes, 
however, only a flash weld requiring special equipment would be ac- 

374 C. E. IVES AND C. J. KUNZ 

ceptable. This difficulty was overcome satisfactorily, however, by 
the adoption of the construction shown in Fig. 3/. Tees are formed 
by butt welding short lengths of Vs-hich inside diameter tubing to 
the manifold tube by the usual gas welding procedure, using a high- 
columbium rod. All openings are inspected and cleared, if necessary, 
prior to welding the manifold to the supply tube 5. Finally, the 
distributing pipes are inserted into the tees and fastened with a small 
amount of silver solder. The outside diameter of the distributing 
tube is almost as great as the inside diameter of the stems in the tee 
so that only a fine line of the silver solder is exposed to the action of 
the developing solution (Fig. 3/). 


1 CLARK, W.: "Standard Development," Phot. J., 49 (Feb., 1925), pp. 76-89. 

1 CRABTREE, J.: "Directional Effects in Continuous Film Processing," J. Soc. 
Mot. Pict. Eng., XVIII (Feb., 1932), p. 207. 

CRABTREE, J., AND WADDELL, J. H.: "Directional Effects in Sound Film Proc- 
essing II," J. Soc. Mot. Pict. Eng., XXI (Nov., 1933), p. 351. 

3 CRABTREE, J. I., AND SCHWINGEL, C. H.: "The Effect of Aeration on the 
Photographic Properties of Developers," /. Soc. Mot. Pict. Eng., XXXIV (April, 
1940), p. 375. 

* HICKMAN, K. C. D., AND HvNDMAN, D. E. : "Plastic Cellulose in Scientific 
Research," J. Frank. Inst., 207 (1929), p. 231. 

6 MIYAGI, O. : "Theory of the Air-Lift Pump with Special Reference to the Slip 
of Air Bubbles in Water," Tech. Reports Tdhoku. Imp. Univ., 4, No. 2 (1924), pp. 
1-18. (Discussed in Bull. No. 84, Nat. Res. Council, "Hydrodynamics" (1932), 
p. 304.) 

O'BRIEN, M. P., AND GOSLINE, J. E. : "Velocity of Large Bubbles in Vertical 
Tubes," Ind. Eng. Chem., 27 (Dec. 1935), p. 1436. 

BROADWELL, B. E., AND WERKING, L. C.: U. S. Pat. 1,988,478 (Jan. 22, 1935). 
WERKING, L. C.: "Fabricated Porous Carbon," Trans. Electrochem. Soc., 74 

(1938), pp. 365-374. 

HATFIELD, M. R., "Fluid Flow through Porous Carbon," Ind. Eng. Chem., 31 
(1939), pp. 1419-1424. 



Summary. The object of this investigation was to ascertain the feasibility of 
using air as a means of agitating developing solutions by determining the effect of 
bubbling air through various commercially used developers on their photographic 

Unseasoned elon-hydroquinone developers of relatively high alkalinity (pH 10.0 
to 10.5} showed a rapid decrease in activity after aeration for l l / 3 hours while elon- 
hydroquinone-borax developers of low alkalinity (pH 8.4 to 8.8} showed increased 
activity (due to the liberation of alkali resulting from oxidation) which then remained 
constant for prolonged periods. 

In general, the alkalinity of developers containing hydroquinone increased on 
aeration, while those containing only elon showed little change. 

Practical tests with processing machines equipped with air agitation devices have 
shown that very constant developing conditions can be maintained when suitably re- 

It is well known that many types of photographic developers tend 
to lose their developing power as a result of exposure to the air. This 
is especially true in the case of developers containing an appreciable 
quantity of pyrogallol which change rapidly from colorless to brown- 
ish red solutions especially when the temperature is much higher 
than 70F. 

Developers of the elon-borax type, however, show no visible signs 
of change under the above conditions but their developing properties 
are more or less affected. 

With hydroquinone developers containing sodium carbonate or 
caustic soda, and which have been allowed to age in the presence of 
air, it is necessary to develop for a longer time than with fresh de- 
velopers in order to secure a given degree of contrast. 1 In 1929, 
Carlton and Crabtree 2 discovered that a developer containing borax 
as the alkali showed increased activity when allowed to age under 
ordinary conditions. Liippo-Cramer 3 and Rzymkowski 4 have also 

* Communication No. 672 from the Kodak Research Laboratories; pre- 
sented at the 1938 Spring Meeting at Washington, D. C.; revised manuscript 
received March 22, 1940. 

** Eastman Kodak Co., Rochester, N. Y. 



shown that sodium sulfite solutions containing developing agents but 
without alkali show increased developing action after being subjected 
to aerial oxidation. 

Mees and Piper 5 encountered aerial oxidation fog with hydroqui- 
none developers and also stated that the oxidation products of py- 
rogallol, when present in developers, caused fog. Similarly, it was 
stated by Crabtree 6 that the oxidation products of developing agents, 
produced by improper methods of mixing, may cause development 
fog. This observation was not confirmed in the later work of Dundon 
and Crabtree 7 when it was found that the addition of oxidized de- 
veloper to a fresh developer tended to decrease fog. This inability 
to produce oxidation fog with existing chemicals was attributed to the 
fact that chemicals manufactured previous to 1920 which readily 
gave fog, must have contained impurities which, on oxidation, be- 
haved as powerful fogging agents. 

Fuchs 8 also found that the oxidation of hydroquinone developers 
by air in contact with an emulsion produced a latent image fog, while 
the oxidation products in the developer tended to decrease rather 
than increase fog. He considers that aerial oxidation fog is the re- 
sult of a latent light image produced by chemi-luminescence which, 
in turn, is a result of oxidation of the developer in contact with the 
film. It is more probable, however, that aerial fog is a result of the 
formation of peroxides which are known to be powerful fogging agents. 

In order to obtain uniform development of a photographic image, it 
is necessary to employ some means whereby fresh solution is constantly 
applied at the surface being developed, otherwise uneven develop- 
ment is obtained. Probably the simplest procedure for developing 
short lengths of film or plates is to develop in a shallow open tray. 
Agitation of the solution is accomplished either by rocking the tray 
or by brushing the emulsion surface with a camel's hair brush. These 
conditions are conducive to rapid oxidation owing to the large sur- 
face of liquid which comes into contact with the air. 

In the case of processing machines in photofinishing establishments, 
the film wetted with developer is exposed to the atmosphere for at 
least 10 per cent of the total developing time while, with certain 
motion picture developing machines, somewhat similar conditions 
prevail which present ideal conditions for aerial oxidation. 

Aerial oxidation also presents serious difficulties when motion 
picture film is developed on a reel when, due to the prolonged contact 
of the wet film with air, it is usually necessary either to add a desensi- 


tizing dye or an increased quantity of sodium sulfite to the solution 
in order to prevent aerial fog. 

Many motion picture laboratories depend upon the circulation 
of the developer as a means of agitation and conditions are often 
such that during recirculation the solution is allowed to overflow into 
a tank situated on the floor below. This overflowing of frothing de- 
veloper with entrapped air furnishes excellent conditions for aerial 

The importance of agitation of the developer in order to obtain 
uniformly developed images is becoming more fully appreciated. 9 
Of the various possible methods of agitation, the use of a stream of 
gas bubbles is effective and economical. 10 Nitrogen gas is to be pre- 
ferred for this purpose, but is expensive. 

The object of this investigation was to ascertain the feasibility of 
using air as a means of agitating developing solutions by determining 
the effect of bubbling air through various commercially used de- 
velopers on their photographic properties. 

Experimental Methods. The following data were obtained by the 
use of a ! 7 /8-inch by 70-inch glass tube closed at one end and mounted 
vertically for holding the solutions through which the air was bubbled. 
Two liters of solution were used for each test which filled the tube for 
about one-half the height, the remaining tube portion serving to 
prevent loss of solution by spattering and frothing. 

The air was passed in at the bottom of the tube and the rate of air 
flow was approximately 5 cubic feet per hour at a temperature of 
72 to 74F which produced a degree of aeration somewhat more 
rigid than that which usually occurs in practice. The air employed 
was filtered and then washed so as to bring it .to near saturation and 
thereby minimize evaporation of the developer liquid. All connec- 
tions, stoppers, and stop-cocks were of glass so as to eliminate any 
possible contamination. 

On the completion of a bubbling test, the two-liter sample of de- 
veloper was withdrawn from the tube and used immediately for 
developing sensitometric strips which had been exposed previously 
on a Type 116 sensitometer. 11 The temperature during development 
was maintained at 65F. When a positive type of developer was 
being tested, Eastman motion picture positive film (1301) was used 
and, for a negative type of developer, Eastman motion picture nega- 
tive film (1217} was employed. 



All pH measurements were made on a slide comparator using La- 
Motte indicators. 

Effect of Aeration with an Elon-Hydroguinone Motion Picture 
Positive Developer. Fig. 1 shows the effect of prolonged air agitation 
on an elon-hydroquinone developer (Formula D-16) with respect to 
(a) the change in density obtained with equal exposures, (6) the 
change in solution alkalinity, and (c) the emulsion fog density values. 
Data are included for (a) fresh D-16, (6) partially exhausted D-16 
(175 ft per gallon), and (c) D-16 containing double the normal 
quantity of sulfite. 

pM II 

o- i 


FIG. 1. Effect of prolonged aeration on an elon-hydroquinone 
developer (Formula D-16). 

Fresh D-16 Developer. The results indicated that with a fresh de- 
veloper aeration caused marked changes in its developing properties. 
At first the density values showed a slight decrease with increasing 
tune of aeration, but after a period of I 1 /? hours, the decrease was 
quite rapid. After 4 ! /2 hours the solution possessed no developing 
properties; the speed loss followed closely the trend in density, and 
the alkalinity increased from a pH of 10.2 to 11.3. 

The increase in developer alkalinity and decrease in activity may 
be explained partially by the probable reaction between hydro- 
quinone, sodium sulfite, and oxygen, with the formation of hydro- 


quinone sulfonates together with 1 mole of sodium hydroxide for 
each mole of hydroquinone reacting, as indicated by the following 
equation : 


The first published accounts of the formation of hydroquinone sul- 
fonates in a developer were by Andresen 12 and Bogisch. 13 A large 
number of investigators have since examined the nature of the 
reaction products of development including Pinnow 14 who, in 1913, 
isolated both hydroquinone monosulfonate and disulfonate. Subse- 
quently, many authors have offered various explanations for the 
mechanism of the formation of these compounds during development 
including Rzymkowski, 4 Lehmann .and Tausch, 15 Seyewetz and 
Szymson, 16 and James and Weissberger. 17 

For the present consideration, the most important fact is that 
sodium hydroxide is formed as a by-product of the oxidation of hydro- 
quinone in the presence of sulfite to hydroquinone monosulfonate. A 
portion of the monosulfonate probably reacts in the same way as 
hydroquinone to form the disulfonate with the liberation of an addi- 
tional quantity of sodium hydroxide, according to the following equa- 


/\SO,Na /NsOjNa 

+ O 2 + 2Na 2 SO 8 -* NazSO* + + NaOH 

\y SO,Nal x/ J 


The by-products of oxidation with a hydroquinone developer are 
therefore sodium sulfate, hydroquinone mono- and disulfonates, and 
sodium hydroxide. Sodium sulfate in the concentration formed 
should have little or no effect on development and, according to 
Evans and Hanson, 18 it is doubtful whether the sulfonates play any 
part in the development of an image. The total effect, therefore, is 
to raise the pU. or degree of alkalinity of the developer which, in turn, 
accelerates the rate of development but this effect is offset by de- 
pletion in the quantity of active developing agents as a result of 

380 J. I. CRABTREE AND C. H. SCHWINGEL [j. s. M. p. E- 

oxidation. During the first stages of aeration, the alkali accumulation 
is sufficient to offset the depletion of developing agents but since the 
activity with increase in pH rapidly levels off, the photographic effect 
of depletion of the developing agents becomes more and more ap- 
parent with prolonged aeration. 

Seasoned Developer. The curves in Fig. 1 also show the effects pro- 
duced by aeration of a developer which had been seasoned previously 
with 175 feet per gallon of flashed 35-mm motion picture positive film. 
The experimental conditions were identical with those for the fresh 
developer. The rate of loss of density was less than when the fresh 
developer was employed, no appreciable change occurring until after 
an intial period of iy 2 hours of aeration, after which the density 
values decreased but not as rapidly as with the fresh developer. 
The pH of the solution showed only a slight increase. 

These observations are quite different from those with fresh de- 
veloper when the pH increased from 10.2 to 11.3. 

No precise explanation can be offered for the difference in behavior 
between the fresh and seasoned developers but it is reasonable to 
suppose that the developer reaction products served as anti-catalysts 
for the oxidation reaction. 

When silver bromide is developed to metallic silver with a hydro- 
quinone-carbonate developer, the following reaction probably takes 
place : 


Na 2 CO, + 2AgBr + Na 2 SO, = + NaHCO, + 2Ag + 2NaBr 


This equation would indicate that a seasoned developer should be 
of slightly lower alkalinity than one in which no film has been de- 
veloped because of the exchange of sodium carbonate for sodium bi- 
carbonate and the elimination of the slightly alkaline salt, sodium 

When hydroquinone is oxidized in a developer by air, the solution 
becomes more alkaline but, when it is oxidized by virtue of perform- 
ing useful photographic work, the solution becomes less alkaline as 
the hydroquinone is depleted. It seems reasonable, therefore, to 
assume that the seasoned developer contained a small quantity of 
sodium bicarbonate which did not materially change the alkalinity of 


the solution but was capable of reacting with its equivalent amount 
of sodium hydroxide formed during oxidation. The equivalent of 
sodium carbonate resulting from the reaction would therefore not 
increase the alkalinity as much as an equivalent quantity of sodium 

During aeration, therefore, the alkalinity of a seasoned developer 
containing hydroquinone would show only a slight initial increase 
until all of the sodium bicarbonate formed during seasoning had re- 
acted with its equivalent of sodium hydroxide. After this reaction, 
the sodium hydroxide formed by further aeration would cause a ma- 
terial increase in the alkalinity of the solution. This would explain 
why the alkalinity of a seasoned developer does not increase at the 
same rate as that of an unseasoned developer during aeration. 

With the seasoned developer, the change in density observed from 
tests during the first 30 minutes was much less pronounced than with 
the unseasoned developer. It is known that small amounts of oxi- 
dized developer, when added to a fresh developer, produce antifogging 
and slight desensitizing effects. 19 This may account for the rapid 
depression in density during the first 30 minutes of treatment using 
the fresh developer. 

A Modified D-16 Developer. Because of the recognized protective 
action of sodium sulfite in retarding the oxidation of organic develop- 
ing agents, it was thought that a higher sulfite content might provide 
a considerable retarding effect during aeration. Therefore, the D-16 
developer containing double the recommended quantity of sodium 
sulfite was prepared and tested. From the results shown in Fig. 1, 
it will be seen that the increased quantity of sodium sulfite produced 
the following effects : 

(2) The effective emulsion speed was increased slightly. 

(2) The change in density values with aeration closely paralleled those with 
the unmodified exhausted developer. 

(5) The pH of the solution was maintained at a relatively constant value. 
(4) No increase in fog was observed. ( 

From these tests, therefore, it is apparent that a positive type of 
developer containing considerably more than the normal quantity of 
sulfite would be somewhat more satisfactory if the developer is to be 
agitated by means of air. 

Results Obtained with a Borax Type of Negative Developer (D-76). 
Data relating to fresh and partially exhausted developers of this type 



are shown in Fig. 2. The curves indicate that changes occurred im- 
mediately after subjecting the developer to aeration and that there 
was no great difference in behavior between fresh and partially ex- 
hausted solutions. Values for gamma and density showed a progres- 
sive increase and, after 20 hours of continuous agitation, gamma 
values were increased approximately 25 per cent and the density 
values for equal exposures increased approximately 40 per cent. 
This differential gain in density and gamma produced a 10 per cent 
increase in speed during the first hour of aeration. Fog values in- 
creased only slightly, changing from 0.10 to 0.16 for the 20-hour test. 





FIG. 2. Effect of aeration of a borax-type of negative developer 
(Formula D-76). 

The alkalinity of the solutions increased uniformly from a pH 
value of 8.6 to 9.2. These observations were identical for fresh and 
seasoned developers. The increase shown in developer activity 
may be attributed to this increase in alkalinity which, undoubtedly, 
proceeded at such a rate as to more than offset the reduction in con- 
centration of the active developing agents. 

There was no appreciable difference in the shape of the H&D 
characteristic curves when comparisons were made at constant gamma 
with fresh developer and with developer which had been subjected 
to aeration. 

April, 1940] 



Results Obtained with the Borax-Boric Acid Negative Developer 
(Formula D-76d). Considering the results obtained with a borax 
developer, it was thought that the borax-boric acid developer devised 
by Carlton and Crabtree 2 might suffer less change on aeration be- 
cause of the tendency of the solution to retain a constant />H value, 
and data with such a developer are given in Fig. 3. 

The curves indicate that the boric acid stabilized the solution 
slightly but had no very pronounced effect until after several hours of 
aeration. The changes in gamma and density for fresh and seasoned 
developers were essentially similar to those obtained with the regular 
borax developer up to a period of 10 hours. 

5 0.8 

2 a * 



4O PT./OAL. 








FIG. 3. Effect of aeration of a borax-boric acid negative developer 
(Formula D-76d). 

The initial change in pH was less abrupt than with the plain borax 
developer and, after 10 hours, the tendency was to remain constant. 

The alkalinity of the developer did not increase at the same rate as 
in the case of the borax developer without boric acid, due to the fact 
that the borax-boric acid combination has a decidedly greater buffer 
action than borax alone for the range of developer alkalinity under 
consideration. Also, the relative increase in alkalinity of either of 
the borax types of developer was much less than in the case of the 
carbonate developer which is accounted for by the fact that a com- 



bination of borax and boric acid or borax alone has a relatively greater 
buffering effect than sodium carbonate. 

Results Obtained with an Elon-Borax Developer Containing a Desen- 
sitizer (Formula D-89}. This developer was of the borax type but 
contained pinakryptol green as a desensitizer and no hydroquinone. 
From Fig. 4 it will be seen that this developer behaves somewhat 
differently from the other types of borax developers. The pH value 
of the solution remained constant at a value of 8.8 throughout the 
tests which would substantiate the previous deductions that it is the 
hydroquinone constituent of a developer which reacts to give in- 
creased alkalinity. The data show that for a fresh developer, the 
gamma and density values decreased continuously from the start in 
contrast to the negative types of borax developers already discussed 











40 FT./GAU. 



FIG. 4. 

Effect of aeration of an elon-borax developer containing a 
desensitizer (Formula D-89}. 

containing hydroquinone, in which case the density and gamma values 
increased during the initial stages of aeration. 

With the seasoned developer, the general effects obtained were 
similar but more pronounced than with the fresh solution. 

Fresh and Seasoned Elon-Borax Developers without Pinakryptol 
Green. Because of the marked difference in behavior between de- 
veloper D-89 and the previous borax developers, it was thought that 
the desensitizer might have some catalytic effect upon the rate of 
oxidation of elon. A solution was prepared, therefore, which con- 
tained no desensitizer. The fresh developer gave practically no 
change in gamma or density, while the seasoned developer produced a 
marked falling off in gamma and density values, the results obtained 

April, 1940] 



being almost identical with those from a fresh developer containing 
pinakryptol green. 

From these results it appeared probable that both pinakryptol 
green and colloidal silver which is formed during exhaustion of de- 
velopers with a high sulfite content catalyzed the oxidation of elon. 

Results Obtained with Developers of the MQ Series. It is evident 
from the preceding experiments that hydroquinone and elon behave 
very differently with respect to oxidation when compounded in de- 
veloping solutions. To check these observed differences further, it 
was thought desirable to test developers having widely differing elon 

FIG. 5. Effect of aeration of developers containing various pro- 
portions of elon and hydroquinone. 

to hydroquinone ratios, the other constituents remaining constant. 
Several members of the MQ series 20 of developers were therefore 
tested. The results of tests with developers MQ-0, MQ-50, and 
MQ-100 are shown in Fig. 5. The developer MQ-0 contained hy- 
droquinone (5 grams per liter) and no elon, MQ-50 contained equal 
quantities of elon and hydroquinone (2.5 grams of each per liter), 
and MQ-100 contained elon (5 grams per liter) and no hydroquinone. 

All the developers contained identical quantities of sodium sulfite 
(75 grams per liter), sodium carbonate (25 grams per liter), and po- 
tassium bromide (1.5 grams per liter) . 

With the MQ-100 developer, gamma and density values did not 


change as the aeration was prolonged, and the alkalinity remained 
constant. These results confirm previous observations with the borax 
developer (D-89], and it was concluded that elon in the presence of 
sulfite in alkaline solution was relatively stable toward oxidation pro- 
vided no catalysts such as colloidal silver or pinakryptol green were 

The data from the tests with MQ-0 also confirmed the previous 
observations that hydroquinone was very susceptible to oxidation in 
the presence of alkaline sulfite solutions and reacted in a manner as 
to increase the alkalinity of the solution. The data with MQ-0 were 
similar to those obtained with D-16 as would be expected, since the 
hydroquinone-elon ratio oiD-16 is relatively high. 

The results showed that the drop in density and gamma values 
with aeration was greatest with MQ-0 and least with MQ-100. Like- 
wise, the alkalinity increase for MQ-50 was intermediate between 
that for MQ-0 and MQ-100. 

Tests with Miscellaneous Developers. Tests with caustic, borax- 
caustic, and borax-carbonate types of developers showed that the 
susceptibility to oxidation of developers containing equivalent elon- 
hydroquinone ratios, with equal quantities of sodium sulfite, was de- 
pendent largely upon the developer alkalinity and not upon the spe- 
cific alkali employed. 

It is true that developers may have equal oxidation susceptibilities 
although their rates of oxidation may be different. This is illus- 
trated by the results obtained with a caustic and a carbonate de- 
veloper containing equal quantities of sodium sulfite and equal hy- 
droquinone-elon ratios. The alkalinity of the solutions was adjusted 
by varying the quantity of alkali so as to give equal pH. values. On 
aeration, both solutions commenced to oxidize at about the same time 
but, as the aeration was continued, the developer containing the 
caustic oxidized at a greater rate than that containing carbonate. 
By measurement it was found that the alkalinity after aeration was 
greatest for equal degrees of aeration in the case of the caustic de- 
veloper and this may be explained by the buffering action of the 
sodium carbonate which tends to maintain a constant pH value as 
caustic is added as a result of oxidation of the hydroquinone. 

The difference in behavior between two developers of the same 
formula DK-40 compounded with (a) Kodalk and (6) an equivalent 
quantity of carbonate is shown in Fig. 6. Although the values for 
the fresh developers were equal, it is seen that the pH of the carbon- 

April, 1940] 



ate solution (7.5 grams per liter) increased at a greater rate than that 
of the corresponding Kodalk developer (20 grams per liter) due, un- 
doubtedly, to the greater buffering action of the Kodalk. The 
gamma values also diminished at a greater rate with the carbonate 
developer, due to the greater rate of oxidation at the higher H value. 

With the exhausted developers, the change in rate of photographic 
activity was somewhat greater than with the fresh developers, con- 
firming the data in Figs. 1 and 2. 

In the above test, the developer was aerated by rotating a small 
reel in a trough containing the developer at a rate of 30 revolutions 





FIG. 6. Showing the difference in behavior between two de- 
velopers of the same formula, DK-40 compounded with (a) Kodalk 
and (b) an equivalent quantity of sodium carbonate. 

per minute. The reel was wrapped with a sheet of Kodaloid so as 
to create a greater surface for aeration. 

From the above data it may be considered that for equal pH values 
of the original developer, for a minimum change in photographic ef- 
fect with aeration, the various alkalies are to be preferred in the order 
of their buffering ability, namely, (a) borax and Kodalk, (6) carbonate, 
and (c) caustic soda. 

Effect of Sodium Sulfite Concentration on the Relative Oxidation 
Rates of Elon and Hydroguinone. In order to determine the change 
in rate of oxidation of elon and hydroquinone in combination with 



varying concentrations of sodium sulfite, various developers were 
compounded according to the following formula : 

Developing agent 
Sodium sulfite (desiccated) 
Sodium carbonate (desiccated) 
Potassium bromide 

5.0 grams 

Varying concentrations 
25.0 grams 
1.5 grams 

The end point of the aeration test was taken as the time required to 
render the solution incapable of developing an image density for a 
given exposure on positive film, when developing for 15 minutes at 

FIG. 7. Effect of sulfite concentration on the rate of oxidation 
of developers containing (a) elon, (6) hydroquinone, and (c) mix- 
tures of elon and hydroquinone. 

65F. From Fig. 7 it is seen that elon was protected by sodium 
sulfite to a much greater degree than hydroquinone. The results 
with mixtures of elon and hydroquinone seem to indicate that the 
rate of oxidation of elon is somewhat accelerated by the oxidation 
products of hydroquinone, especially at low sulfite concentrations. 

Effect of Temperature on Rate of Developer Oxidation. Tests with 
Formula D-16 indicated that a change in temperature from 65 to 
95F had only a slight effect on the rate of oxidation. 

Removal of Developer Sludge by Aeration. Developing solutions, 
especially those for use with positive types of film, are usually dis- 


carded when they become colored and tend to stain the gelatin. 21 
It was observed that when a badly discolored D-16 developer was 
agitated with air it became lighter in color and the opalescence in 
the solution disappeared. On closer observation, it was seen that 
the froth which formed on the surface of the solution was quite dark 
in color and could be skimmed off, thereby leaving the solution rela- 
tively clear. Also, the clearing effect was most complete after about 
15 minutes of agitation when the developer was pale straw colored 
but, if aeration was continued further, the hue darkened to a very 
dark brown. Possibly practical use could be made of this observa- 
tion as a means of clarifying large quantities of developer. 

Agitation with Nitrogen. When nitrogen was used for agitating 
various developers, no change was observed in their photographic 
properties even after 10 hours of continuous aeration. 

Effect of Carbon Dioxide. Since air contains approximately 0.045 
per cent (by weight) of carbon dioxide, which gas reacts with alkalies 
to form bicarbonates, thereby lowering their degree of alkalinity, it 
might be expected that carbonation of the alkalies would reduce their 
photographic activity apart from the oxidation phenomena. 

Accordingly, aqueous solutions containing 20 grams per liter of 
sodium carbonate and 40 grams per liter of Kodalk, respectively, were 
bubbled with carbon dioxide for 16 hours in which time an appreciable 
effect would be obtained if air were used. The solution was then 
used for compounding the D-16 and DK-40 developers. No essential 
difference in the properties between these developers and solutions 
prepared with the treated alkalies was detected, showing that the 
degree of carbonation produced by the aeration was very slight. 

Calculations would indicate that only 0.0015 mol of carbon dioxide 
would pass per hour when bubbling air at the rate of 5 cubic feet per 
hour. To eliminate the effect of carbon dioxide the air could, of 
course, be passed through a soda lime or sodium hydroxide absorber. 

Effect of Sodium Ar senile and Sodium Hypophosphite. It was 
thought that other protecting agents, such as sodium arsenite and 
sodium hypophosphite might produce better stability in a developer 
than sodium sulfite and, by so doing, lessen the tendency of the de- 
veloper toward oxidation. When a portion of the sodium sulfite in a 
developer was replaced by equivalent quantities of sodium arsenite 
and sodium hypophosphite (from mere traces to the limit of solu- 
bility of the salts) no improved protection over that furnished by the 



sodium sulfite was observed although the gamma-time characteristics 
were changed. 

Effect of Developer Oxidation on Graininess. Throughout the tests 
discussed above, the effect of developer oxidation products on the 
graininess of the developed images was constantly observed. When 
comparisons were made of images developed to equal densities and 
gammas at equal effective emulsion speeds, in solutions which had 
been aerated to different degrees, no marked differences in graininess 
were noted as a direct result of the aeration. 

Practical Considerations. In order to determine the practicability 
of employing air for agitation purposes when developing photographic 
materials, it is necessary to know approximately the relative extent 





FIG. 8. Effect of exhaustion of D-16 developer, with and without 
aeration, when in use in a motion picture processing machine. 

of the developer exhaustion produced by (a) aeration and (6) exhaus- 
tion by virtue of performing useful work in developing the image. 

It is known that the magnitude of (b) is quite considerable but this 
can be compensated for by suitable replenishing 22 ' 23 with the result 
that the photographic properties of a developer can be maintained 
over prolonged periods of use. 

In order to obtain a measure of the relative magnitude of (a) and 
(6) in practice, exhaustion tests were made with the D-16 developer 
in a continuous processing machine, using motion picture positive 
film. The capacity of the tank was 25 gallons, the rate of recircula- 
tion 5 liters per minute, and the film traveled at the rate of 25 feet 
per minute. 

A test run was made over a period of 10 hours without aeration 
and without replenishing, and the test repeated with aeration and 


without replenishing. A study of the curves in Fig. 8 shows that 
the rate of change in gamma and density with exhaustion was 
practically identical in both cases, showing that the aeration had 
little or no effect on the photographic properties of the solution. 
When the developer was merely aerated (without exhaustion or re- 
plenishing) no visible change in photographic activity occurred over a 
period of 10 hours. 

The method of air agitation was essentially that described by Ives 
and Kunz 10 although the severity of the agitation was much less than 
that obtained by bubbling air through a tube as used in the tests 
described above. Also, with the machine aerator, the very small air 
bubbles which formed in the tube aerator were eliminated, thereby 
reducing the rate of oxidation for a given degree of agitation. 

It is apparent, therefore, that under the conditions described, the 
effects produced by aeration are negligible as compared with those 
produced by exhaustion. 

In practice, with the above machine using D-16 and motion picture 
positive film, replenishing is accomplished by adding developer at the 
rate of 225 cc per minute when developing at the rate of 25 feet per 
minute, the excess overflowing into the drain. The fresh developer 
in the tank is compounded two-third normal strength but the normal 
developer is used for replenishing. In this way, any slight initial in- 
crease in activity by virtue of aeration is compensated for by the lower 
activity of the weaker solution. Under these conditions, for a run 
of 200,000 feet of film, 400 gallons of replenisher were added. In the 
case of very dense prints, the replenisher flow was increased to 275 
cc per minute while, with sound records (no picture record), this was 
reduced to 125 cc per minute. 

The D-76 developer has also been used successfully in combination 
with aerial agitation. In this case, when the capacity of the tank 
was 40 gallons, one-half strength D-76, but with the normal quantity 
of sulfite (100 grams per liter), was used at the start and full strength 
D-76 used for replenishing. At a developing rate of 15 feet per 
minute, the flow of replenisher was 600 cc per minute, and the aver- 
age life of the developer approximately 100 feet per gallon. 

The effect of constitution on stability against change by aeration 
is also illustrated by the case of a low alkalinity slow-working de- 
veloper for variable density sound negatives (developer A , Table III). 
This formula had performed satisfactorily in a commercial laboratory 
previous to the time when an attempt was made to use it under con- 

392 J. I. CRABTREE AND C. H. SCHWINGEL [j. s. M. P. E. 

ditions where agitation of the developer was obtained by the use of 
compressed air. 

When fresh, the developing time to produce a gamma of 0.35 at 
70F with Eastman sound recording negative emulsion 1359 was 12 
minutes. After a few days of intermittent use during which sound 
negatives were developed with agitation by compressed air, it was 
found that the developing rate increased to such a degree that the 
time for a gamma of 0.35 was reduced to 5 or 6 minutes. This in- 
crease in developer activity was more rapid in the later stages and 
could not be anticipated exactly. 


Feet per 

Developer (Cumulative) 






It was found that the increase in activity was accompanied by an 
increase in alkalinity and that the original developing rate could be 
restored by the addition of 0.25 gram of citric acid per liter. None of 
the change could be ascribed to the replenishment because this con- 
sisted merely of adding developer of the original formula at the rate 
of one gallon for 200 feet of film processed. The increase in activity 
in the course of use was largely a result of oxidation of the hydro- 
quinone by air with the production of sodium hydroxide. Under 
the new condition of use the rate of aeration must have been increased 
relative to the rate at which reaction products of development were 
produced. The buffering effect of citric acid in such a formula would 
not be very great at the H of 8.5 for the freshly mixed developer. 

After some small scale tests, developer B (Table III) was com- 
pounded in order to insure more stable operation under the same con- 
ditions of use. 

Minutes of 

Time of 
for Gamma = 0.3.3 

























April, 1940] 






00 _) 

N 1C d O 








TH 1C Oi CO 

^ H-) 


d <M' d do 

d ^ 



o o o >c 



TH Tj< O O 

o^ 4 



(M TH 



o o o ic 


1C 1C 1C TH 



t^ <N 



1C 1C O O 1C 


C^l N 1C 1C TH 



" t^ IN 



O O O iC 


1C 1C >C TH 


'fe * 

a % 3 2 


w t> JS 

rJ g -^ 



id d _i 


< % 8 * 

H 9 s > 
* 2 Q 

O O 

00 00 _j 





O O 


ci id d 


TH CO 00 


CO O CO t^ 00 CO 


O CO OS 00 O O 

TH "- 1 






cj *O 



'S % CD 

CO ^H 

a -SJ "2 

84&J 1 


H 3 CS C "O 
3 GO O C -TH 

O 1 C c .5 cS 
O C ft oo 

.^ d 

3 1 
s -a 


^ Tj Tj 00 O 

^ "O iS irt fj '^ 

cS cs y 'M * ^ 

o o o .S o ,* 

394 J. I. CRABTREE AND C. H. SCHWINGEL []. S. M. p. E 

Results obtained with developer formulas A and B under similar 
conditions of use in a 120-gallon quantity are shown in Table I. 
With each, sound negatives were developed from time to time 
throughout a period of several days. The emulsion for testing was 
Eastman sound recording negative 1359. Development times are 
given for a gamma of 0.35. 

Thus the necessary stability shown by a practically constant time 
of development was obtained with somewhat increased chemical 
concentrations. The chemical consumption would depend upon the 
extent of use before rejection or upon replenishing rate as determined 
over a longer period of use than that shown. 

In contrast to the case just described unintentional aeration of a 
comparatively stable developer may produce very troublesome 
changes in developer performance. The following data show the 
change in pH. of a D-76 type of developer in the course of a few hours 
use in a system in which the air was being entrained in the return pipe 
of the recirculation line as a result of allowing the pipe to run only 
partly filled. 


Period of Use Feet of Film Developed 

(Cumulative) (Cumulative) pH 


45 min. 200 8.5 

1'A hours 2400 8.7 

2 s A hours 4400 8.8 

3 / hours 5200 8.9 

7"/4 hours 5900 9.0 

10 3 A hours 6700 9.1 

This change in pH was found to accompany an increase in the rate 
of development of about 60 per cent. 

There are various ways in which this aeration can occur as, for ex- 
ample, at a leaky pump packing. The air passing through a pump is 
probably broken up into small bubbles, providing a large surface of 
contact with the developer. A similar rate of change in pH and ac- 
tivity was obtained by delivering air in very fine bubbles into a 500-cc 
quantity of the same developer. Under these conditions the rate of 
change was about the same either at atmospheric pressure or at ap- 
proximately two atmospheres. An almost identical change in the 
photographic characteristics was brought about by the addition to the 
fresh developer of sufficient sodium hydroxide to obtain the same in- 


crease in pH as that caused by the aeration. This, however, would 
not be true for a case where the aeration had been prolonged to the 
point where developing agents were largely consumed by oxidation. 


(1) The effect of agitation of developers with air has been studied 
under conditions somewhat more severe than usually occur with 
motion picture film processing machines. 

(2) Fresh elon-hydroquinone developers of relatively high al- 
kalinity (pH = 10.0-10.5) oxidized more rapidly than those of low 
alkalinity (borax type, pH. = 8.4-8.8). Developers of the first type 
showed a very rapid decrease in activity after iy 2 hours of aeration 
while those of the second type generally showed increased activity 
with no falling off after prolonged treatment (15 hours) . 

(3) In general, partially exhausted elon-hydroquinone developers 
showed less susceptibility to aeration than fresh developers. 

(4) Developers with high sodium sulfite content (50 to 100 grams 
per liter) were found to be generally more stable than developers con- 
taining less than 50 grams per liter. 

(5) The susceptibility of a developer to oxidation is dependent 
upon the initial alkalinity of the solution, other constituents remain- 
ing constant, and not upon the particular alkali employed. For a 
given alkali, the rate of oxidation of a developer increased with in- 
creasing alkalinity. In the case of hydroquinone this rate has been 
shown by other investigators to vary as the square of the hydroxyl 
ion concentration. 24 

(6) On aeration, the alkalinity of all developers containing hydro- 
quinone increased, while those containing only elon showed no change. 

(7) The effect of temperature on the oxidation reactions taking 
place in elon-hydroquinone developers was negligible between tem- 
peratures of 65 and 75 F. 

(8) Discolored and sludged developers may be partially clarified 
by aeration. 

(9) Agitation with nitrogen over a long period produced no 
change in the activity of a developer. 

(Iff) No apparent change in graininess characteristics of the 
developers tested was produced by aeration. 

(11) Experience has shown that it is entirely practical to agitate 
motion picture machine developers with air when suitably replenished. 
The method 10 is economical and, with suitably constructed grids for 


distributing the air, excellent uniformity of development can be ob- 

The authors are greatly indebted to C. E. Ives and C. J. Kunz 
for assistance in the experimental work. 


1 CRABTREE, J. I.: "Development of Motion Picture Film by the Reel-and- 
Tank Systems," Trans. Soc. Mot. Pict. Eng. (1923), No. 16, p. 163. (Revised 

1 CARLTON, H. C., AND CRABTREE, J. I.: "Some Properties of Fine-Grain 
Developers for Motion Picture Film," Trans. Soc. Mot. Pict. Eng. (1929), No. 38, 
p. 406. 

3 Lflppo- CRAMER: "Aging of Metol-Hydroquinone Solutions," Phot. Ind., 29 
(1930), p. 1180; ibid., 29 (1931), p. 548. 

4 RZYMKOWSKI, J.: "The Increase of Development Velocity with Age of 
Weak Alkaline Developers," Phot. Ind., 29 (1931), p. 825. 

MEES, C. E. K., AND PIPER, C. W. : "The Fogging Powers of Developers," 
Phot. J., 35 (1911), p. 226; ibid., 36 (1912), p. 221. 

CRABTREE, J. I. : "Chemical Fog," A mer. Ann. Phot., 33 (1919), p. 20; Brit. 
J. Phot., 66 (1919), p. 97. 

7 DUNDON, M. L., AND CRABTREE, J. I.: "The Fogging Properties of De- 
velopers," Trans. Soc. Mot. Pict. Eng. (1928), No. 36, p. 1096. 

8 FUCHS, E. : "Aerial Oxidation of Developers and the Formation of Fog," 
Phot. Ind., 27 (1924), p. 56. 

9 CRABTREE, J. : "Directional Effects in Continuous Film Processing," J. 
Soc. Mot. Pict. Eng., XVH (Feb., 1932), p. 207; CRABTREE, J., AND WADDELL, 
J. H.: "Directional Effects in Sound-Film Processing II," J. Soc. Mot. Pict. 
Eng., XXI (Nov., 1933), p. 351. 

10 IVES, C. E., AND KUNZ, C. J.: "Solution Agitation by Means of Com- 
pressed Air," J. Soc. Mot. Pict. Eng., XXXIV (April, 1940), p. 364. 

11 JONES, L. A.: "A Motion Picture Laboratory Sensitometer," /. Soc. Mot. 
Pict. Eng., XVH (Oct., 1931), p. 536. 

11 ANDRESEN, M. : "The Chemistry of Organic Developers," Phot. Korr., 37 
(1900), p. 185. 

" BOGISCH, A.: "Reduction and Developing Power," Phot. Korr., 37 (1900), 
p. 93. 

14 PINNOW, J.: "The Action of Oxygen on Hydroquinone and Sodium Sul- 
fite," Z. Wiss. Phot., 11 (1913), p. 289; "The Deterioration with Keeping of 
Sulfite-Hydroquinone Solutions," Z. wiss. Phot., 22 (Aug., 1922), p. 72; "Sulfite 
in Developers," Phot. Rund., 60 (1923), p. 27. 

u LEHMANN, E., AND TAUSCH, E.: Phot. Korr., 71 (1935), p. 17; TAUSCH, E.: 
"The Chemistry of Photographic Developers," Dissertation, Berlin, 1934. 

16 SEYEWETZ, A., AND SZYMSON, S. : "On the Oxidation of Organic Develop- 
ing Agents," Bull, de la Soc. Fran, de Phot., 21 (1934), p. 71. 

17 JAMES, T. H., AND WEISSBERGER, A.: "Oxidation Processes, XI The 
Autoxidation of Durohydroquinone, " /. Amer. Ghent. Soc., 60 (1938), p. 98. 


18 EVANS, R. M., AND HANSON, W. T., JR.: "Reduction Potential and Photo- 
graphic Developers," "The Effect of Sulfite in Developer Solutions," /. Phys. 
Chem., 41 (April, 1937), p. 509. 

19 DUNDON, M. L., AND CRABTREE, J. I.: "Investigations on Photographic 
Developers III," "The Effect of Desensitizers in Development," Trans. Soc. 
Mot. Pict. Eng. (1926), No. 26, p. 111. 

20 MEES, C. E. K.: "A Simplified Method of Writing Developing Formulas," 
Brit. J. Phot., 64 (1917), p. 535. 

21 CRABTREE, J. I., AND DUNDON, M. L. : "The Staining Properties of Motion 
Picture Developers," Trans. Soc. Mot. Pict. Eng. (1926), No. 25, p. 108. 

22 CRABTREE, J. I., AND IVES, C. E. : "A Replenishing Solution for a Motion 
Picture Positive Film Developer," J. Soc. Mot. Pict. Eng., XV (Sept., 1926), p. 

23 JAMES, T. H., SNELL, J. M., AND WEISSBERGER, A.: "Oxidation Processes, 
XII The Autoxidation of Hydroquinone and of the Mono-, and Di-, and Tri- 
methylhydroquinones," J. Amer. Chem. Soc., 60 (1938), p. 2084. 

24 EVANS, R. M. : "Maintenance of a Developer by Continuous Replenish- 
ment," /. Soc. Mot. Pict. Eng., XXXI (Sept., 1938), p. 273. 


Summary. Motion picture film and the characteristics of its support are 
discussed from the point of view of safety. A survey is given of the work carried out 
by the Eastman Kodak Company in conjunction with the motion picture industry, 
boards of underwriters, the National Fire Protection Association, and government 
bodies, on the control of the fire hazard in the production, distribution, and use of 
cellulose nitrate film for motion pictures. The characteristics and use of cellulose 
acetate film are considered in relation to the problem. 

It is a privilege and an honor to be asked to present the story of the 
accomplishments in fire prevention in the motion picture industry. 
The Eastman Kodak Co., having developed and introduced the 
flexible, transparent film that made-the motion picture industry possi- 
ble, and because of its long experience in the manufacture of this film, 
is undoubtedly in the best position to outline this problem in its en- 

To understand the magnitude of the problem and the difficulties in- 
volved in fire prevention in the motion picture industry, we must go 
back to the beginning of the use of film in photography. In 1880, 
when Mr. Eastman began the manufacture of dry plates, he realized 
that photography could not be made popular with the average indi- 
vidual until something less bulky and less breakable than glass-plate 
negatives could be produced. 

In 1885, he first introduced stripping film, as it was called, which 
consisted of paper, coated first with a layer of soluble gelatin, then 
with a thin layer of collodion (nitrocellulose), and finally with a coat- 
ing of sensitized emulsion. The film was exposed by the photog- 
rapher, then sent back to the Company, where it was developed and 
laid emulsion-side down on a glass plate. The soluble gelatin layer 

* An address delivered before the Greater New York Safety Council, Hotel 
Pennsylvania, New York, N. Y., March 27, 1939. 

** Eastman Kodak Co., Rochester, N. Y. 


was then softened and the paper was stripped off, thus leaving a 
transparent negative on the glass plate for printing. 

Although this next development, stripping film, broadened the field 
of amateur photography considerably, Mr. Eastman realized that it 
was not the final answer. He therefore employed a chemist in 1886 
to devote all of his time to the'development of a suitable flexible, non- 
breakable, transparent material which could replace the paper of the 
stripping film and also serve as a permanent support for the sensitized 

Collodion, which is a solution of cellulose nitrate in ether and alco- 
hol, had been used in wet-plate photography, the method that re- 
placed the early Daguerreotype process. It was the best known and 
practically the only known transparent material that would form 
itself into a continuous film when the solvents were evaporated. The 
film formed from collodion was found wanting, however, in many re- 
spects as a support for photographic emulsions. 

Mr. Eastman was familiar with collodion from his experiments 
with wet-plate photography before he developed the Eastman gelatin 
dry plate with which he began his business career in photography. 
His later experiments in search of a transparent, flexible emulsion- 
support demonstrated the unsuitability of collodion for this purpose. 
Despite the defects of collodion, however, it was natural that the ex- 
perimenter should concentrate on nitrocellulose, its main ingredient, 
in his search for a transparent, flexible support. After 3 years of 
experiments, a formula of nitrocellulose and camphor in solution, 
suitable for negative support, was perfected, and commercial produc- 
tion of nitrocellulose film base was started in 1889. 


Nitrocellulose is inflammable. Inflammability increases with the 
degree of nitration of the cellulose. Experiments demonstrated that, 
to produce a nitrocellulose suitable for film support, a low degree of 
nitration must be maintained so that the resulting nitrocellulose 
would be less inflammable, less subject to decomposition, and of the 
correct solubility. 

Nitrocellulose as used for photographic purposes is not explosive, 
but is inflammable ; and intelligent care must be exercised in its pro- 
duction, storage, and use. Nitrocellulose film base will not only burn 
rapidly when ignited, but also, if subjected to sufficient heat, will 
decompose without flame. This film base contains enough combined 

400 A. F. SULZER LF. s. M. P. E. 

oxygen to maintain decomposition when once started, even in a limited 
air supply. Decomposition liberates comparatively large quantities 
of carbon dioxide, carbon monoxide, and oxides of nitrogen, which 
under certain conditions are dangerous to life. Some of these lib- 
erated gases are also inflammable, and under some conditions are 

Since decomposition, when once started, will maintain itself and 
generate enough heat to produce combustion, the obvious control, 
lies, not in methods intended to smother the fire by excluding the air 
supply, but by the application of large quantities of water sufficient to 
cool the burning film below the decomposing temperature of about 
300 degrees F. Experiments demonstrated that such an application 
of water will also cool the liberated gases sufficiently to make explo- 
sion, and even ignition, unlikely. This theory was followed in fire- 
protection measures developed by the Eastman Kodak Co., to which 
we will refer later. 

Nitrocellulose film was first used only in roll film for amateur 
photography. The development of equipment for the taking and 
subsequent projection of pictures which would seem to move had 
been waiting, however, for a flexible film. Therefore, almost im- 
mediately after Mr. Eastman announced the availability of flexible 
negative material, Thomas A. Edison, realizing that this was the an- 
swer to one of his most perplexing problems, sent one of his men to 
Rochester to bring back samples with which he might carry on his 
experiments on motion picture cameras. Mr. Edison's first confirm- 
ing order, covering a prior delivery of motion picture film, bears the 
date of Sept. 2, 1889. 


Motion pictures first appeared commercially in peep shows in 1894. 
Then, on May 20, 1895, the first motion pictures were projected on a 
screen commercially, at 153 Broadway, New York City. This was a 
four-minute picture of a prize fight. Progress was at first slow, but, 
by the late nineties and early nineteen hundreds, motion picture 
shows often referred to as nickelodeons were becoming more 
common. Production was low, however, and the quantities of film 
on hand were necessarily small. 

These early films were short. A show usually consisted of one, or 
at most two, subjects. The show lasted twenty to thirty minutes. 
Twenty to thirty shows were put on daily ; but the seating capacity 


of the theaters, many of which were remodeled stores, usually did not 
exceed one hundred. 

Because of the constantly increasing use of motion picture film, 
which is not backed or interleaved with paper, and because of some 
serious fires that occurred in theaters and exchanges, the Eastman 
Kodak Co. became vitally interested in the problem of fire prevention. 

In 1906, Kodak began experimenting with cellulose acetate, which 
has the same transparent properties as cellulose nitrate and in addi- 
tion is no more inflammable than paper, wood, or many other forms 
of ordinary cellulose. In addition to its lower inflammability, cellu- 
lose acetate will not decompose readily when heated ; and, except for 
carbon monoxide, it does not give off toxic gases when it burns. No 
more carbon monoxide is released from cellulose acetate when it is 
burned in a limited air supply than is given off by equal quantities of 
ordinary cellulose, such as paper or wood. 


In 1909, Eastman had developed cellulose acetate to a point where 
the Company felt it could be substituted for nitrocellulose in motion 
picture film. To give effect to this development, Mr. Eastman ar- 
ranged a meeting with the leaders of the motion picture producing 
companies. Because the advantages of the new film were obvious to 
all, little argument was needed to reach an agreement whereby only 
cellulose-acetate film was to be supplied by the Company thereafter. 

Experience demonstrated, however, that acetate film was not as 
strong mechanically as nitrocellulose film and that it became brittle 
with use. Difficulty in the projection of the acetate film was experi- 
enced partly because of the inferior quality of the film but also be- 
cause of the inferior projection equipment of that day and the rough 
handling to which the film was subjected. 

Although some improvements ensued in film, in projection equip- 
ment, and in handling, the motion picture producers asked in 1911 to 
be released from their agreement to use only cellulose-acetate film. 
Thus, in less than two years, the Eastman first attempt to substitute 
slow-burning cellulose-acetate film for nitrocellulose film came to an 
end. This attempt failed, not because of lack of cooperation on the 
part of the motion picture producers, but because of the failure of the 
cellulose-acetate film to perform satisfactorily under the conditions 
to which it was subjected. 

The period from 1911 to 1922 was one of research, education, and 

402 A. F. SULZER IJ. S. M. P. E. 

cooperation : research in methods of making the production, distri- 
bution, and exhibition of motion pictures safe to the public and the 
workers involved ; education of everyone involved in these activities, 
including not only the industries themselves but also the fire depart- 
ments, transportation companies, and public officials, local, state, and 
national; and, finally, complete cooperation which resulted from 
these research and educational undertakings. 

With the return in 1911 to the production and use of nitrocellulose 
film for the motion picture industry, Eastman cooperated whole- 
heartedly with the motion picture producers, the national and local 
boards of underwriters, the National Fire Protection Association, 
and the various governmental bureaus and administrators, in the de- 
velopment of devices and methods to control the fire hazard in the 
production, distribution, and use, of cellulose-nitrate film for motion 

Although in a period of five years 1912 to 1917 the reports of the 
N. Y. City Fire Department show that film was the cause of only 
12 /ioo of one per cent of the number of fires in N. Y. City, and 
28 /ioo of one per cent of the losses by fire, the potential hazard in 
the use of the film was realized. As a first official step, ordinances 
were passed, at the instigation of the boards of fire underwriters, to 
make the projection of motion pictures in the theaters safe for the 


Following the Ferguson Building fire in Pittsburgh on Sept. 7, 1909, 
the U. S. Geological Survey made a thorough investigation to deter- 
mine the probable causes. From this investigation, and from labora- 
tory tests, it was concluded that the explosion accompanying that 
fire was caused by the ignition of gases generated under pressure in a 
closed, unvented vault in which a quantity of nitrocellulose film de- 
composed after being ignited by the breaking of an electric light bulb. 
This explosion did not occur in the vault, but the gases which escaped 
into an adjoining room formed an explosive mixture with the air of the 
room and were ignited by a fire burning outside the vault. 

The laboratory tests made at that time confirmed these conclusions 
and proved that nitrocellulose film is not explosive; and proved, 
furthermore, that the gases generated by film decomposition at at- 
mospheric pressure are of a low inflammability, but that, if the film 
decomposes under pressure, the gases generated, when properly 
diluted by air, are explosive. 


The results of these tests were given wide circulation, ancl ordi- 
nances were enacted, in practically all of the major cities, requiring 
the projection rooms in motion picture theaters to be amply vented 
to the open air and to be completely isolated from the theater audi- 
torium. These precautions and other restrictions plus education as 
to the volume of film and its hazards are largely responsible for the 
practical absence of fires in motion picture theaters. 

Another source of hazard lay in the wornout, obsolete, or discarded 
film. With the greatly increased use of motion pictures and with 
longer subjects, this obsolete film piled up rapidly in the exchanges. 
In 1918, Eastman inaugurated the plan of purchasing this discarded 
film, for recovery of its constituent materials for non-photographic 
purposes. The film was, and still is, collected and shipped to Kodak 
Park, Rochester, and to other responsible converters, thus removing 
one of the greatest sources of fire hazards in the exchanges. In addi- 
tion, safe methods were devised for handling this scrap, and these 
were made available to others who wished to carry on the recovery of 
scrap film as a business, with safety to property and life. 

Despite the fact that the number of fires and losses attributable to 
nitrocellulose film continued small in comparison with other causes 
such as gasoline, matches, smoking, and carelessness, agitation 
against the use of nitrocellulose film continued. 

In 1915, the committee on explosives and combustibles of the 
National Fire Protection Association, in cooperation with the N. Y. 
City Fire Department and the Universal Film Co., conducted a test 
at Fort Lee, N. J., by burning a large quantity of discarded motion 
picture film in a vault which, though properly vented, was not 
equipped with automatic sprinklers. 

This test was very spectacular. No explosion accompanied the 
fire, but the heat was so intense that a giant torchlike blast of flame 
shot horizontally out of the vault vent for many feet. Numerous 
tests which had been made by Eastman in its work of protecting its 
employees and its own property from the hazards of film fires, had 
demonstrated, previous to the Fort Lee tests, that properly arranged 
sprinklers will control film fires. 

Immediately following the Fort Lee tests, the Eastman Company, 
believing that the severity of fires under conditions of the Fort Lee 
tests could be greatly lessened and could be controlled, ran a series of 
tests in 1915 and 1916 to determine the inflammability of film, the 
protective effectiveness of water in varying quantities from properly 

404 A. F. STJLZER [J. S. M. P. E. 

arranged sprinklers, and the protective effectiveness of various 
methods of packaging and storing motion picture film. 

In the Eastman tests, conditions as to volume and arrangement of 
film were equal to or more severe than the conditions in the Fort Lee 
tests. These tests, together with earlier tests run by the Company, 
proved conclusively that, with properly constructed and vented 
vaults, film fires can be readily extinguished by sprinklers, and, in 
addition, decomposition can be prevented from communicating "to 
other films in the same storage racks. These tests also proved that, 
in properly vented vaults, sprinklers will cool the liberated gases so 
that explosion, or even ignition, of liberated gases is improbable. 

These tests were witnessed by officials of the N. Y. City Fire De- 
partment and also by representatives of the underwriters and insur- 
ance companies. The results did much to convince these people that 
film fires in vaults can be controlled by automatic sprinklers, with 
proper venting and with proper limitation of quantity stored; and 
also that the safe storage of large quantities of film is possible if proper 
precautions are taken. 

While the Eastman Company was experimenting with methods of 
fire prevention for nitrocellulose film, the inspection department of the 
Associated Factory Mutual Fire Insurance Companies was carrying 
on similar experiments with pyroxylin plastics, commonly known as 
Celluloid. A comprehensive report of these tests was published in 
1916. Pyroxylin contains nitrocellulose similar to that used in 
motion picture film. The findings in this report were in agreement 
with the findings of the U. S. Geological Survey in the investigation of 
the Ferguson Building fire in Pittsburgh. This latter report, how- 
ever, included definite specifications for limitation of storage-vault 
capacity and for adequate venting and sprinkling. 

In 1919, the N.F.P.A.'s committee on explosives recommended a 
similar code, or specification, for the storage of cellulose-nitrate film. 
These reports, and those of the Eastman experiments in the same 
years, dealt with the fundamentals, and have formed the basis, first, 
for fire underwriters' rulings, and, later, for laws and ordinances 
governing the transportation, storage, and handling of all nitrocellu- 
lose motion picture film. 

From 1916 to 1919, Eastman prepared a series of booklets entitled, 
Suggestions on Fire Prevention. The first booklet dealt with auto- 
matic sprinklers; the second, with housekeeping; the third with 
motion picture film, its characteristics, and hazards; and the fourth 


with the results of the tests on motion picture film fires in vaults. 
Arrangements were made with the motion picture producers to make 
these booklets available to all persons in the industry responsible for 
the production, processing, handling, and storage, of film. These 
booklets, except one describing the tests, were written in plain, non- 
technical language, so that their message could be readily understood 
by the non-technical employees in the industry. 


In addition to this prepared material, the Eastman offer to send 
out experts to all exchanges to inspect the exchanges and instruct the 
managers in proper fire-prevention methods was accepted by the 
motion picture producers. Six men were specially trained for this 
work and covered the four hundred exchanges in the U. S. and Can- 
ada. Formal and very complete reports of these inspections were 
forwarded to the exchange managers and also to the officials at the 
headquarters of the companies owning the exchanges. 

Following the first inspections, the Eastman Company, collaborat- 
ing with the Fire Underwriters and with various government agencies, 
drew up plans and specifications for film exchange buildings. These 
plans and specifications were so prepared that they could be readily 
adapted to local conditions, and they were made available to motion 
picture producers and to others who wished to build new film ex- 
changes or to rehabilitate existing ones. Eastman also provided a 
consulting service for the producers in this work. 

Follow-up inspections showed that conditions had materially im- 
proved, indicating that, if those vitally interested in the problem of 
fire prevention are informed, effective cooperation is possible. This 
work of inspection and consultation was carried on by Eastman until 
1922, when it was turned over to the newly organized M.P.P.D.A. 
The best evidence that that organization has successfully carried on 
the work is to be found in the comparative absence of exchange and 
theater fires, even with the constantly increasing volume of film 
produced and handled. 

Beginning in 1922 and continuing up to the present time, we have 
had a long period of consolidation of the progress made and of as- 
similation of the information developed by experience and experi- 
mentation. Although as has been pointed out the number of 
fires and the amount of fire loss caused by film, or in which film be- 

406 A. F. SULZER U. S. M. P. E. 

came involved, had been small, compared with numbers and losses 
from other causes, there had been a number of spectacular fires. 


In spite of the fact that investigations following these fires showed 
that, in general, known preventive measures had not been properly 
applied, there were demands periodically to outlaw the use of nitro- 
cellulose film. In 1919, at Ottawa, the resolutions committee of 
N.F.P.A. offered the following resolution: "The universal adoption 
and exclusive use of slow -burning motion picture film with national, 
provincial, state and local legislation to prevent the continued manu- 
facture and distribution of material having the hazardous properties 
of gun cotton stock now commonly employed." 

This was Item 10 in a series of eleven items in the resolutions pro- 
posed by the committee. Item 10 caused more discussion than all 
the other ten items in the proposed resolutions, combined. There 
was a decided difference of opinion; but, because of the extensive 
laboratory tests and practical full-scale tests that had been carried on 
to determine the nature and hazards of cellulose-nitrate film, and be- 
cause of the demonstrated effectiveness of methods of control, the 
great preponderance of opinion favored regulation of use rather than 
imposing the impractical alternative in Item 10 on the great and grow- 
ing industry of production and exhibition of motion pictures. The 
resolutions, when adopted, formed the platform for the year, and were 
used as the basis of insurance-rate rulings and requirements and for 
legislation to make them effective. 


Experience with cellulose-acetate film for commercial motion pic- 
tures in the years 1909 to 1911 had demonstrated that it was entirely 
unsatisfactory. Progress had been made in improving its wearing 
qualities, but in 1919 it still was far behind nitrocellulose film in this 

Legislation in practically all communities had been enacted to 
make the exhibition of nitrocellulose film safe for the public. The 
conditions in the exchanges had been improved, and the Eastman 
Company, at the time of the Ottawa meeting, was preparing to send 
out experts to help the motion picture producers to improve still 
further the conditions in the exchanges. Tests by government 
agencies and by others had proved that nitrocellulose film is not ex- 


plosive as we have pointed out and records proved that the num- 
ber of fires attributable to film was negligible compared with the 
number attributable to any other cause. 

All of these facts and others were brought out in the discussion of 
Item 10 of the 1919 resolutions. As a result, the resolutions as finally 
adopted included a revised Item 10, reading as follows: "That the 
use of motion picture projection machines without a standard booth 
ventilated to the outside of the building, in churches, schools, clubs, 
hospitals and homes, be prohibited unless the film used is of the slow- 
burning type and that state and municipal laws and ordinances be 
adopted regulating motion picture exchanges, tending toward the 
ultimate end that motion picture films of the nitrocellulose type be 
replaced when practicable by a slow-burning film." 

The great majority of the membership of the N.F.P.A. is made up 
of men representing the insurance companies, of fire underwriters, 
and of public officials. Only a small number of members represent 
either the film manufacturers or the motion picture producers. 
When these facts are taken into consideration, the action taken on 
this resolution is convincing evidence of the enlightened cooperation 
which made the control of fire hazard in the motion picture industry 

The M.P.P.D.A. had taken up, shortly after it was organized in 
1922, the educational and inspection work started by Eastman in 
1919, and had accomplished much in the control of use, transporta- 
tion, and handling, of nitrocellulose film. From time to time, how- 
ever, nitrocellulose film appeared in stores and found its way into use 
in improperly protected projectors. In addition, scrap nitrocellulose 
film was in some cases transported and handled in an improper 

Because of these difficulties, the N.F.P.A. public -information com- 
mittee, in its report in 1923, proposed recommending to the states and 
provinces of the United States and Canada the enactment of a model 
law to control the use of nitrocellulose film. This model law provided 
for "the control of use of nitrocellulose motion picture film and for the 
licensing of manufacture, use, handling, disposition, and transporta- 
tion of such film." 

As in the case of the 1919 resolution, this proposed law was thor- 
oughly and earnestly discussed. No one opposed the idea of regula- 
tion or the necessity for such regulation. A minority, however, felt 
that the Association should not pass such a recommendation, but 

408 A. F. SULZER 

should again go on record as supporting the early substitution of slow- 
burning film for nitrocellulose film for all purposes. 

After much debate which again brought out the progress which had 
been made in the control of the fire hazard in the use of nitrocellulose 
film, the Association adopted unanimously the committee's proposed 
model law. This action and the action at Ottawa in 1919 are good 
examples of an association membership made up principally of persons 
not selfishly interested in a commercial enterprise taking constructive 
action to safeguard the interests and well-being of the public, instead 
of destructive action, which could not have been as fruitful in safe- 
guarding the interests of all concerned. 

As further evidence of the M.P.P.D.A.'s interest in the matter of 
pubic welfare, attention should be directed to its reports to the 1924 
and 1925 annual meetings of the N.F.P.A. These are progress re- 
ports giving accounts of cooperation and outstanding accomplish- 
ment. (Cooperation, I must point out again, is the thread that has 
run through the whole fabric.) 

More thorough and more frequent inspections, cooperation with 
and from local fire departments, introduction of fire drills, circulari- 
zation of exchanges with educational matter, sponsoring of bills for 
the control of use of nitrocellulose motion picture film, are some of the 
activities of the M.P.P.D.A. Film boards of trade were organized 
in many cities. These boards included in their membership both the 
members of the M.P.P.D.A. and of the independent companies. 
Thus the organized efforts in accident and fire prevention were ex- 
tended to the entire industry. 

This educational and inspection work has been carried on by the 
M.P.P.D.A. to the present day. The annual cost, although heavy, is 
justified by the results. The increase in volume of film handled has 
been enormous, and the number of persons involved in the many 
necessary operations has increased accordingly. The price of safety 
in any industry is eternal vigilance. 

Naturally, this story told by the Eastman Kodak Co., in spite of 
effort to view it dispassionately, must needs be colored to some extent 
by the Company's interest in the matter. If its actions were said to 
be motivated by self-interest, however, it was at the very least an 
enlightened self-interest, and the same is no less true of the motion 
picture industry as a whole. As I said in my opening remarks, it is a 
story of accomplishment of great value to the public, made possible 
by one dominant factor cooperation. 


During the Conventions of the Society, symposiums on new motion picture ap- 
paratus are held, in which various manufacturers of equipment describe and demon- 
strate their new products and developments. Some of this equipment is described 
in the following pages; the remainder will be published in subsequent issues of the 



Since the early days of sound motion pictures, the service branch of the in- 
dustry has kept pace with the various developments and improvements in the 
art. In 1928 the field engineer was equipped with a minimum of tools and test 
equipment. In contrast, today he carries a complete set of modern service in- 
struments and tools which comprise the following : 

(1) Complete technical data on Photophone and equipment of other manu- 

(2) Special Weston analyzer 

(3) Special Weston power level meter 

(4) Socket selector kit 

(5) Tool kit 

(6) Special wrenches 

(7) Speed counter 

(8) Loud speaker adjustment tool 

(9) Standard frequency test-film 

(10) 7000 and 9000-cycle focusing films 

(11) Lateral adjustment film (buzz track) 

(12) Push-pull test-film 

(13) Academy dialog and music film 

(14) Universal a-c bridge for measurement of resistance, capacity, and induc- 

(15) Cathode-ray oscillograph 

(16) Beat-frequency oscillator 

(17) Emergency amplifier and speaker system 

Technical Data. Each field engineer maintains a complete file of Photophone 
equipment bulletins and, in addition, is furnished with up-to-date technical 

* Presented at the June 21, 1939, Meeting of the Atlantic Coast Section. 
** RCA Manufacturing Co., Camden, N. J. 



information on all new improvements in the art. Complete data on equipment 
other than Photophone are furnished to him and these data are kept up-to-date, 
so that RCA field engineers can service any type of theater sound reproducing 

Analyzer. The Weston analyzer, together with the power level meter and 
socket selector kit, were made to RCA specifications for application to theater 
sound work. The analyzer incorporates a 20,000-ohm-per-volt meter with scales 
up to 1000 volts. In addition, provision is made for checking any range of cur- 
rent or resistance normally encountered in routine service work. The socket 
selector kit is designed for checking tubes in the amplifying equipment making 
it possible to test under dynamic conditions. 

Power Level Meter. Since a 15-ohm output impedance has been standard on 
Photophone equipment for a number of years, the power level meter is calibrated 
for this impedance and a 12.5-milliwatt reference level. Charts are provided 
from which correction factors can be obtained when the meter is used on cir- 
cuits of different impedance or when other reference levels are necessary. This 
meter is used in conjunction with the standard frequency film or beat-frequency 
oscillator in obtaining overall system frequency response or complete transmission 

Tool Kit. The tool kit contains all the necessary tools for proper installation 
and service operation on the complete equipment. In addition there are special 
sound-head wrenches and motor alignment tools. A Starrett speed counter is 
included for accurately measuring film speed. 

Frequency Test- Film. The present frequency test-film has been designed with 
the view toward making it more useful in field work. Accordingly, identical 
tracks are recorded on each edge of the film, thereby eliminating the need for re- 
winding after each test. This greatly speeds up the work of taking response 

Thirty-three frequencies are included, from 30 to 10,000 cycles, with the 1000- 
cycle reference at the beginning and end of each track. Additional frequency re- 
cordings are included between 2000 cycles and 3000 cycles to provide a more 
comprehensive overall response test. The response is held within 0.5 db 
throughout the frequency range. 

Buzz-Track and 9000-Cycle Film. For adjustment of film position with respect 
to the sound-head light-beam, each engineer carries a small section of so-called 
buzz-track. The recording consists of two narrow chopper tracks so spaced that 
neither will affect the light-beam if the guide rollers are in proper lateral adjust- 
ment. This test-film has been in use since its development, in 1930, by W. W. 
Jones, now Manager of New York District Service Operations, to whom the 
patent was originally issued. 

On the other edge of the film there is a 9000-cyde recording employed for focus- 
ing the light-beam on the film. The film is used in conjunction with an output 
meter to determine when the beam is correctly focused on the sound-track. 

Push-Pull Test- Film. With the advent of push-pull sound-heads, it was 
necessary to provide a test-film for adjustment of the optical systems. This 
film consists of a 6-mil "septum" track on one edge for adjusting the division of 
the light-beam and a 300-cycle track on the other edge, for correctly balancing the 


output of the dual photocell. The 300-cycle portion is also used to balance the 
output from each sound-head. 

Theater Sound Test-Reel. RCA field engineers are now equipped with the 
latest theater sound test-reel produced by the Research Council of the Academy 
of Motion Picture Arts and Sciences. The following description of the film is 
taken from the Academy Technical Bulletin: 

"The reel contains sound and picture, the sound consisting of dialog and music 
recordings so chosen that the assembled reel contains a representative example 
of sound as currently recorded by each sound department. One of these record- 
ings is a 'Hi-Range' print which serves as a check on the amplifier capacity in 
relation to the volume of the auditorium under consideration. 

FIG. 1. Universal Inductance, Capacity, and Resistance Bridge. 

"The reel also contains approximately 100 feet of piano and 12 feet of 3000- 
cycle recordings included for the purpose of furnishing a more critical flutter test. 

"After setting the theater sound reproducing equipment to the Standard 
Electrical Characteristic, the Theater Sound Test-Reel furnishes a tool by which 
an optimum setting for presence and intelligibility, combined with a natural 
balance between high and low frequencies, may be obtained for all current 
product and for the individual theater. 

"The use of this reel demonstrates the inadvisability of having too much low- 
frequency electrical response which brings out noise-reduction bumps, foot- 
steps and parasitic low-frequency noises present on the set. 

"It might be pointed out that judgment is required in the use of the Theater 
Sound Test-Reel, as the product must be evaluated in terms of the material at 
hand, that is, crowd noises and people talking in loud voices or excited manners 



should not be expected to have the same quality and chest tones which are pres- 
ent in conversational dialog in a quiet, intimate scene. 

"The Research Council and the Committee have always felt that electrical and 
acoustical curves furnish valuable means of setting equipment, but that the final 
criteria should be critical listening tests of the equipment. For this reason all 
Theater Sound Reproducing Equipment Standards to date have been set up on the 
basis of listening tests correlated with engineering data. 

"One of the purposes of the Standard Electrical Characteristic is to provide a 
basis for an eventual standard recording characteristic. We believe that the 
new Theater Sound Test-Reel demonstrates the fact that the recording charac- 

FIG. 2. Cathode-Ray Oscillograph. 

teristics of the various studios are very much closer together than they were a 
year or two ago. 

"The material contained in the reel is not necessarily a sample of the best re- 
cording available but is typical of the average. 

"In an average theater, set to the Standard Electrical Characteristic, the reel 
will play through entirely upon one fader setting (with, of course,' the exception 
of the Hi-Range print sequence, for which the fader must be raised 6 db)." 

Universal A-C Bridge. To enable the field engineer to measure inductance and 
capacity as well as resistance, the RCA universal bridge (Fig. 1) is employed. 
This instrument is invaluable for checking reactors and transformers for shorted 
turns or other trouble, and for accurately determining values of capacitors. The 
measuring ranges are 100 nh to lOh, lO/upf to 10/if, and 1 ohm to 1 megohm. For 


higher values of resistance, the test analyzer is employed, since this has ranges up 
to 10 megohms. 

The instrument consists of a variable-ratio-arm Wheatstone bridge having 
three standards each of inductance, capacity, and resistance. A vacuum-tube 
1000-cycle oscillator and a two-stage amplifier, together with their power supply, 
make up the major part of the equipment. The only additional equipment re- 
quired is a "null" indicator, for which the power level meter is employed. Power 
is obtained from any 110 to 120- volt, 25 to 60-cycle source. The complete 
instrument weighs only 6 pounds. 

Cathode-Ray Oscillograph. Another device which has been in general use for 
over three years and has proved extremely useful in theater service, is the cathode- 
ray oscillograph (Fig. 2). This instrument is probably the most versatile device 
yet developed for the study of radio and audio-frequency phenomena. 

FIG. 3. Portable Beat-Frequency Oscillator. 

By means of the oscillograph the field engineer can quickly localize sources of 
hum in the sound systems and check hum patterns, determine where distortion is 
introduced, check phasing of networks, and perform many other routine checks 
which up to a few years ago were impossible to do. This instrument weighs 21 
pounds complete and is entirely self-contained, requiring only a source of a-c 
power supply. 

Beat- Frequency Oscillator. For quickly checking the audio-frequency re- 
sponse and power output of a theater sound system, the RCA portable beat- 
frequency oscillator (Fig. 3) is provided. This instrument is extremely rugged, 
weighs only 15 pounds, and is remarkably stable for its size. The hum level is 
60 db below maximum output. 

Besides its use in checking audio amplifiers, it is also extremely valuable for 
determining source of buzzes, rattles, etc., in stage surroundings and auditorium 
fixtures. It is also used as a routine check on the loud speaker systems to be 
sure that the units are free from distortion. 



April, 1940] 



Emergency Amplifier System. In the latter part of 1936, RCA introduced to 
exhibitors a complete portable emergency sound system, so designed that it 
would be possible to keep the show running if every piece of equipment were 
down except one sound-head and projector (Fig. 4). This system proved so 
valuable that more than 150 are now located in theaters throughout the country, 
and besides this nearly every RCA field engineer carries one as standard equip- 

The most recent design incorporates many improvements for greater ease and 
simplicity in operation. The amplifier and all controls are housed in a small 
metal cabinet, with all necessary cables for making connections to sound-heads 
and stage lines. Sufficient audio power is available to carry even the largest 
theater, and connections can be made to any type of sound equipment regardless 
of manufacture. 


FIG. 5. Audio Curve Tracer, simplified circuit. 


While the various pieces of standard equipment already described are usually 
adequate for routine service work, quite often problems arise which require 
specialized test equipment. RCA is continually at work developing new test 
equipment to handle these problems, as well as new equipment and methods to 
simplify and speed up routine tests. 

Audio Curve Tracer. One of the recently developed instruments in this classi- 
fication is the audio curve tracer. This is a portable device which traces an am- 
plifier response curve automatically on the screen of a cathode-ray tube. By 
using a tube with a long-persistence screen, such as the RCA -9 10, the image is 
retained long enough for it to be studied, photographed, or a second curve super- 
imposed on it for comparison. With this instrument it is possible to run an 
accurate frequency response curve in approximately thirty seconds. 



Fig. 5 is a schematic drawing showing the operation of the instrument. The 
output of a beat-frequency oscillator is fed both into the "horizontal control" 
potentiometer R-l and into the input of the amplifier under test. The voltage 
developed across R-l is fed through a resistor-capacitor network to ground. The 
characteristics of this network are such that with a constant voltage R-l impressed 
across the network, the voltage developed across R-6 varies directly as the loga- 
rithm of the frequency impressed. This is shown in the small curve above. The 
voltage E-6 is impressed across a diode section of the RCA-6H6 tube and is recti- 
fied. The rectified voltage causes current to flow through resistor R-7 producing 
a negative bias voltage at the grid of the RCA-57 d-c amplifier tube. The greater 
the voltage E6 becomes the more negative the bias on the RCA-57 becomes. 
The plate supply voltage to the d-c amplifier is 450 volts. The actual voltage at 


FIG. 6. Overall schematic diagram of complete Audio Curve Tracer. 

the plate of this tube is equal to this voltage minus the voltage drop in the plate 
resistor R-8 caused by the flow of plate current. Hence, as the plate current is 
reduced the drop across the plate resistor is reduced and the voltage at the plate 
of the tube becomes higher. The plate voltage is applied directly to one of the 
horizontal deflection plates of the cathode-ray tube. The opposite deflection 
plate is at some potential above ground which is determined by the setting of the 
centering control potentiometer R-3. 

When the beat-frequency oscillator is set at a low frequency the voltage E6 is 
low. Hence the current through R-7 is small and therefore the value of negative 
bias is small. This causes a large plate current to flow in the RCA-57 and reduces 
the voltage at the plate of the tube because of the large drop in the plate resistor. 
This reduces the voltage applied to the right deflection plate below that at which 

April, 1940] 



the left is set causing the beam to be deflected to the left (that is, toward the more 
positive plate) . 

When the frequency of the oscillator is raised the voltage E6 goes up, more 
current passes through R-7 producing a higher bias voltage, the plate current 
of the RCA -57 is reduced, and the drop across R-8 is very small, leaving the volt- 
age at the plate of the tube and consequently at the right deflection plate higher 
than the voltage of the left deflection plate. Under this condition the negative 
beam of electrons is deflected toward the right plate which is now the more posi- 

Thus by merely tuning the beat-frequency oscillator through its frequency 

FIG. 7. Audio Curve Tracer. 

range the electron beam is moved across the screen horizontally with a displace- 
ment which is always proportional to the logarithm of the frequency impressed. 

The output of the amplifier is fed through the "vertical control" potentiome- 
ter R-2 to ground. A portion of the output voltage is rectified by the second 
diode in the RCA-6H6, amplified by a d-c amplifier identical to the one already 
described and applied to the vertical deflection plates. By analysis similar to that 
for the horizontal deflection circuits, it is evident that an increase in amplifier 
output will cause an upward deflection of the beam. 

Thus, for any given frequency at which the oscillator may be set, the frequency 
determines the horizontal position of the spot, and the output level of the ampli- 



fier determines the vertical position of the spot. Therefore, if the oscillator is 
swept through its frequency range the spot will trace a response curve of the 
amplifier under test. Fig. 6 shows the overall schematic diagram of the complete 
audio curve tracer. 

In addition to the beat-frequency oscillator as a source of signal, a continu- 
ously variable frequency test-film is available which can be run through the sound- 
head to produce an overall response curve of the complete sound system. 

The "film-oscillator" switch at the input of the horizontal amplifier connects 
either of two high-frequency boosters in the circuit. In the "oscillator" position 
a slight rise is added to make up for the reduction in output of the oscillator at 
the high frequencies. A more pronounced rise is provided in the "film" position 
to compensate for the optical losses in the sound-head at the high-frequency end. 

An additional control, the "30-cycle adjustment," has been added so that the 

FIG. 8. Flutter Indicator. 

voltage drop at the plate of the horizontal d-c amplifier can be lowered inde- 
pendently of the voltage at the vertical d-c amplifier, making it possible to move 
the spot horizontally without affecting its vertical position. 

Fig. 7 is a picture of the audio curve tracer. This is housed in a carrying case 
similar to that of the three-inch oscillograph. The front panel contains the 
"focus" and "intensity" controls, the "horizontal" and "vertical" inputs, the 
"horizontal gain" and the "vertical gain" controls, the "horizontal amplifier 
gain" control and the "film-oscillator" compensation switch. The "centering 
control" and the "30-cycle adjustment" control are screw-driver adjustment con- 
trols on the side of the case. 

Flutter Indicator. Early in 1933 RCA introduced the rotary stabilizer sound- 
head, which revolutionized sound reproduction by minimizing to a large extent 
the flutter which was characteristic of the earlier types of sound-heads. How- 


ever, since there are still thousands of older equipments in use which are subject to 
nutter trouble, RCA has available a Flutter Indicator which will aid the service 
engineer in making adjustments to reduce nutter to a minimum (Fig. 8). 

Flutter is actually frequency modulation of the reproduced tone caused by 
irregular motion of the film past the scanning beam; hence any one of several 
frequency-discriminating circuits may be used to detect these irregularities. 

In the past few years, several different types of flutter-measuring devices have 
been built. Most of these employ circuits which are quite similar to those used 
in automatic frequency-control circuits used in broadcast receivers, or frequency- 
deviation meters used in transmitting stations. While these instruments are 
extremely accurate, they require the use of one or more vacuum-tubes together 
with associated power supply circuits. This reduces the portability and hence 
makes the units unsuitable for field work. 

The discriminating network used in this instrument is merely the familiar 
Wheatstone bridge used as an impedance bridge with reactive impedances in two 
of the legs. The constants have been chosen so that when a 3000-cycle signal is 
fed into the network it can be balanced, and a meter placed across the network 
will read zero. If the frequency of the input signal varies above and below 3000 
cycles, the network will become unbalanced by an amount that is proportional 
to the variation and the meter will show a corresponding deflection. 

A specially recorded 3000-cycle film with low flutter content is run through the 
sound-head under test and the flutter indicator is connected to the output of 
the system amplifier. Any variations in output can be measured directly on the 

The input transformer has several impedance taps, so this instrument can 
properly terminate the system amplifier without the use of additional load re- 
sistors. An input control provides vernier adjustment of the input level to the 
measuring network. The "read-calibrate" switch allows the meter to be used 
for calibrating the input to the measuring network or for reading the flutter 
voltage developed across the output of the network. Two ranges of sensitivity 
are made available through the use of the range selector switch. These ranges 
are 0.5 per cent full scale or 2 per cent full scale. Resistance and capacity balance 
controls allow the resonance frequency of the measuring network to be shifted 
slightly to compensate for slight variations in speed between various sound- 
heads. The meter is a 5000-ohm rectox type volume indicator with a scale that is 
hand-calibrated directly in per cent flutter. 

By using this instrument as a guide in adjusting the tension in the sound-gates 
of older types of sound-heads, it has been possible to reduce the flutter from as 
much as 2 per cent to as little as 0.3 per cent. 

Scanning Illumination Test-Track. Another new tool, which is available 
through the Academy of Motion Picture Arts and Sciences, to give the field 
engineer a better check on the operation of the theater sound system is the scan- 
ning illumination test-track. 

This film is made up of seventeen consecutive sound-tracks, each of which is 
displaced a different distance from the guided edge of the film. The individual 
tracks are unilateral tracks approximately 7 mils in width and modulated ap- 
proximately 100 per cent at 1000 cycles. The distance between the center lines of 
the consecutive tracks is 6 mils, allowing a slight overlap from one track to the 



[J. S. M. P. E. 

other. The centerline of the first track occurs at approximately 197 mils from 
the edge of the film and the centerline of the last track 292 mils from the edge of 
the film, and the total track width is approximately 110 mils. Each track 
announces itself by number at the start, and the length of steady-state condi- 
tion is 10 feet with 2 feet allowed for moving from one track to the next and 1 
foot allowed for the announcement. This makes a total of \3 feet for each of 
seventeen sections. 

FIG. 9. Sound-Level Meter. 

A single running of this film provides the following information: 

(1) The length of the slit 

(2) The uniformity of illumination across the slit 

(5) The approximate amount of weave in the machine 

As an example of the information obtained from running this film, it was run 
through an older type equipment and the following facts were established: The 
effective length of the slit was approximately 73 mils; the illumination across the 
slit was uniform within * 1 db ; and the weave in the machine was approxi- 
mately 6 mils. 

Sound-Level Meter. While listening tests in theaters are final criteria for good 
sound reproduction, very often acoustic response readings taken in the auditorium 
will indicate where changes are necessary to improve quality. RCA has employed 


the General Radio 759-A sound level meter (Fig. 9) for this purpose since the 
early part of 1937. In addition, it is extremely useful for checking noise levels in 
projection rooms and the theater proper, for checking extraneous noise produced 
by fans in air-conditioning systems, and the effectiveness of vibration insulation of 
power equipment. 

For checking acoustic response of speaker systems, a warble-tone frequency 
reel is employed as a source of signal, to reduce as much as possible the effects of 
standing waves in the theater. The microphone is set up at various points 
throughout the theater, and then readings are taken from 30 cycles to 9400 cycles 
at each station. The plotted curves so obtained indicate approximately the 
acoustic response in various sections of the theater, and this can usually be con- 
firmed by listening tests. 

On the basis of the acoustic response so obtained, adjustments can be made 
for proper sound quality from the speaker system. This procedure has been 
used in numerous theaters and the results as compared to listening tests were 
very gratifying. 

FIG. 10. Vibration Pick-Up. 

This device is also very valuable in checking the sound distribution throughout 
the auditorium. Response curves run at several points quickly show up any 
deficiencies. The speakers can be accurately angled on the basis of such readings 
to give the optimum sound distribution. 


Other test instruments have been developed by various branches of RCA for 
specific applications. These are often used for theater work in routine service 
or solving special problems. One of these is the RCA vibration pick-up (Fig. 10). 

Vibration Pick-Up. This unit, with its associated equipment, is very useful 
in locating defective gears, bearings, or other moving parts. Such defects usually 
show up as "knocks" occurring at regular intervals, or as vibrations at an audio 
frequency. If the approximate location and the frequency of the "vibration" 
can be determined, the exact location of the defect is rather easy to find. 



The output of this vibration pick-up is fed into an amplifier and in turn to the 
vertical plates of a cathode-ray oscillograph. A prod is provided on the vibra- 
tion pick-up for prodding around the sound-head until the approximate source 
of the vibration is located, as indicated by a maximum deflection on the oscillo- 
graph screen. 

If the frequency of this vibration or knock is known the problem is still further 
simplified. The frequency can be determined by using the cathode-ray oscillo- 
graph externally synchronized by an audio beat-frequency oscillator or an RCA 
synchronizing generator. 

FIG. 11. Ultrasensitive D-C Meter. 

If the synchronizing generator is used, it is coupled to the projector crank 
shaft or to a sound-head sprocket shaft. It generates a synchronizing voltage at 
intervals which are directly related to the rpm of the shaft. The synchronizer 
also provides a movable marker voltage which can be impressed on the signal 
under observation and gives a means of marking the oscillograph trace with 
respect to the angular position of the shaft. 

Knowing the rpm of the motor and sound-head sprockets, the problem is still 
further reduced and can be solved by associating the frequency of knock or 


vibration with the rpm of the moving parts in the vicinity of the source of vibra- 

Ultrasensitive D-C Meter. Another instrument developed by RCA for use in 
laboratories and handling special field measurements is the ultrasensitive d-c 
meter (Fig. 11). This is a ruggedly built, portable precision device, for measur- 
ing small values of current and voltage, and a wide range of resistance. 

Current measurements as low as 0.02 microampere and up to 10,000 micro- 
amperes can be made over twelve different scale ranges. D-c voltage measure- 
ments from 0.1 volt to 500 volts over eight scale ranges can be made. Resistance 
values from 0. 1 megohm up to 200,000 megohms can be checked with this instru- 

The instrument consists of a multiplicity of input circuits, a three-stage d-c 
feedback amplifier, and a meter circuit. The amplifier is so designed that the 
meter can not burn out or even deviate in calibration through overload unless the 
sensitivity push-button is held down. 

The sensitivity of the instrument approaches that of the average reflecting 
galvanometer. The overall accuracy for all ranges of current or voltage measure- 
ments is 2 per cent of full scale at ambient temperatures of 50 to 100F and 
normal humidity. For resistance measurements the maximum deflection error 
is 0.1 inch at mid-scale and approaches zero at ends of the scale. 

The instrument is particularly useful in theater sound work for measuring 
photocell currents, leakage currents between tube electrodes, and between cir- 
cuit elements. The unusually high input resistance, five megohms or better, on 
all ranges of the voltmeter circuit enables accurate measurement of d-c voltages 
across high-impedance circuits such as those existing between tube electrodes or 
across circuit elements such as bias-resistors. 




Engineering Vice-President Executive Vice-President 

Editorial Vice-President 



Financial Vice-President 


Convention Vice-President 















Chairman, Mid-West 


Chairman, Atlantic 

Coast Section 


Chairman, Pacific 

Coast Section 


(Atlantic Coast) 

*P. J. LARSEN, Chairman 

D. E. HYNDMAN, Past-Chairman **R. O. STROCK, Manager 

*J. A. MAURER, Sec.-Treas. *H. GRIFFIN, Manager 

(Pacific Coast) 

*L. L. RYDER, Chairman 

]. O. AALBERG, Past-Chairman **P. MOLE, Manager 

*A. M. GUNDELFINGER, Sec.-Treas. *K. F. MORGAN, Manager 

*W. C. MILLER, Manager **H. W. MOYSE, Manager 

*J. DURST, Manager **H. REMERSHEID, Manager 


*J. A. DUBRAY, Chairman 
S. A. LUKES, Past-Chairman *O. B. DEPUE, Manager 

*I. JACX>BSEN, Sec.-Treas. 

*C. H. STONE, Manager 

* Term expires December 31, 1940. 
** Term expires December 31, 1941. 


(Correct to March 25th; additional appointments or changes may be made at any 
time during the year, as necessity or expediency may require.) 




(East Coast) 

G. FRIEDL, JR., Chairman 

(West Coast) 

E. HUSE, Chairman 





J. I. CRABTREE, Chairman 





G. F. RACKETT, Chairman 





W. C. KUNZMANN, Chairman 





F. H. HOTCHKISS, Chairman 




[]. S. M. P. E. 



G. R. GIROUX, Chairman 


A. C. HARDY, Chairman 

C. E. K. MEES 








J. G. FRAYNE, Chairman 



H. E. WHITE, Chairman 


J. A. MAURER, Chairman 





S. HARRIS, Chairman (East Coast) 
L. A. AICHOLTZ, Chairman (West Coast) 









April, 1940] 




E. R. GEIB, Chairman 















R. H. RAY 














New York 








District of Columbia 
































O. F. NEU 

J. S. ClFRE 











A ustralia , 











G. D. LAL 

















New Zealand 





\J. S. M. P. E. 


M. E. GILLETTE, Chairman 





J. G. FRAYNE, Chairman 




N. LEVINSON, Chairman 





J. HABER, Chairman 


P. A. McGuiRE 


H. G. TASKER, Chairman 








D. B. JOY, Chairman 












April, 1940] 








E. C. RICHARDSON, Chairman 


D. B. JOY 


P. C. GOLDMARK, Chairman 





A. N. GOLDSMITH, Chairman 
(Projection Practice Sub- Committee) 
H. RUBIN, Sub-Chairman 






(Theater Design Sub-Committee) 
B. SCHLANGER, Sub-Chairman 





R. F. Ross 


American Documentation Institute. 

Sectional Committee on Motion Pictures, ASA. 

Sectional Committee on Photography, A SA . 
Inter-Society Color Council. 












**A. N. GOLDSMITH, Chairman 


O. F. NEU 

*(G. W. BOOTH) 

E. W. ELY 





D. B. JOY 




* Alternate. 
**SMPE delegates. 



The editors present for convenient reference a list of articles dealing with subjects 
cognate to motion picture engineering published in a number of selected journals. 
Photostatic copies may be obtained from the Library of Congress, Washington, D. C., 
or from the New York Public Library, New York, N. Y. Micro copies of articles 
in magazines that are available may be obtained from the Bibliofilm Service, Depart- 
ment of Agriculture, Washington, D. C. 

Acoustical Society of America, Journal 

11 (January, 1940), No. 3 

Review of Cardioid Type Unidirectional Microphones 
(pp. 296-302) 

Application of Piezoelectric Vibration Pick-Ups to 
Measurement of Acceleration, Velocity and Dis- 
placement (pp. 303-307) 

Loudness Level to Loudness Conversion Chart (pp. 

On the Theory of Fluctuations in the Decay of Sound 
(pp. 324-332) 

Noise and Vibration Isolation (pp. 341-345) 

Sound in the Theater (pp. 346-351) 

American Cinematographer 

21 (February, 1940), No. 2 
Studying Photoelectric Exposure Metering (pp. 64-65), 

Pt. Ill 
New Gadget Coordinates Meter, Makeup, Lighting 

(pp. 67-88) 

Handicaps Against India's Film Production (pp. 
84-85), Pt. I 

British Journal of Photography 

86 (December 29, 1939), No. 4156 
Progress in Colour (pp. 758-760) 

87 (January 5, 1940), No. 4157 
Progress in Colour (pp. 7-8) 

87 (January 12, 1940), No. 4158 
Improved Tri-Chromatic Separation (pp. 15-17) 
Progress in Colour (pp. 17-19) 

Educational Screen 

29 (January, 1940). No. 1 
Motion Pictures Not for Theaters (pp. 16-18), Pt. 15 












U. S. M. P. E. 

Electronics and Television and Short- Wave World 

13 (January, 1940), No. 143 
Construction of Apparatus for Recording Sound on 

Steel Wire (pp. 4-10) 
The Television Range Finder (p. 19) 


1 (December, 1939), No. 6 

Das Zwischenfilmverfahren (Intermediate Film Tele- 
vision) (pp. 201-210), Pt. Ill 

Bruckenmodulationsschaltungen (Bridge Circuits for 
Modulating Purposes) (pp. 211-215) 

Gesichtspunkte zum Bau von Grossprojektionsempf- 
angern (Television Projection Tube Receivers) 
(pp. 216-219) 

Ueber Photozellen mit Sekundarelektronenverviel- 
fachern (Photoelectric Cells with Secondary Emis- 
sion Multiplier) (pp. 226-230) 


21 (December, 1939), No. 12 

Ein abgekurztes Verfahren zur betriebsmassigen Bes- 
timmung des Donnereffektes (A Quick Method for 
Determination of the "Donnereffekt" in the Plant) 
(pp. 263-265) 

Material- und Arbeitsersparnis bei der Nachbearbeit- 
ung von Bildtonfilmen (Saving of Material and 
Labor During After-Treatment of Sound Film) 
(pp. 265-268) 

Arbeit- und filmsparende Umroller f iir die Kinotechnik 
(Film Rewinder for Motion Pictures) (pp. 268-269) 

Bericht uber die Ermittelung der Grossenverhaltnisse 
der deutschen Filmtheater als Grundlange fur die 
Klimatisierung (Report on Relative Size of German 
Motion Picture Theaters as Basis of Air Condition- 
ing) (pp. 269-270) 

Die Oberflache des allernachsten Fixsternes (Surfaces 
of the Newest Fixed Stars) (pp. 271-272) 

Institute of Radio Engineers, Proceedings 
28 (January, 1940), No. 1 

A New Standard Volume Indicator and Reference 
Level (pp. 1-17) 













April, 1940] 



International Photographer 

11 (January, 1940), No. 12 
Are Photometers Necessary? (pp. 6, 8) 
History of Negro Motion Pictures (pp. 16-17) 
Fluorescent Lamp as a Key Light (pp. 14-15) 

12 (February, 1940), No. 1 
Professional Type Exposure Meters (p. 8) 

New High Speed Sheet Film (Eastman Tri-X Pan- 
chromatic) (pp. 21-22) 

International Projectionist 

14 (December, 1939), No. 11 
Three-Dimensional Motion Pictures: A Review and 

Forecast (pp. 19-21) 
Novel Cinetymer Reel Footage Indicator (pp. 22-23) 

Journal of Applied Physics 

11 (January, 1940), No. 1 
Theory of the Photographic Latent-Image Formation 

(pp. 18-34) 

Physics in Color Photography (pp. 46-55) 
The Design of Wide- Aperture Photographic Objectives 
(pp. 56-69) 

Photographische Industrie 

37 (December 20, 1939), No. 51 

Vereinfachte Beleuchtungsberechnung bei der Kino- 
projection (Simplified Illumination Computation in 
Motion Picture Projectors) (pp. 1231-1232) 












Officers and Committees in Charge 

E. A. WILLIFORD, President 
S. K. WOLF, Past-President 
W. C. KUNZMANN, Convention Vice-President 
J. I. CRABTREE, Editorial Vice-P resident 
S. HARRIS, Chairman, Papers Committee 
J. HABER, Chairman, Publicity Committee 
H. GRIFFIN, Chairman, Convention Projection 
E. R. GEIB, Chairman, Membership Committee 
H. BLUMBERG, Chairman, Local Arrangements 

Reception and Local Arrangements 






Registration and Information 

W. C. KUNZMANN, Chairman 


Hotel and Transportation 

E. O. WILSCHKE, Chairman 





J. HABER, Chairman 


P. A. McGuiRE 



Convention Projection 

H. GRIFFIN, Chairman 




Officers and Members Projectionists Local 310, IATSE 

Banquet and Dance 

M. C. BATSEL, Chairman 




Ladies' Reception Committee 

MRS. O. F. NEU, Hostess 

assisted by 




Miss L. A. MOVER, Social Director, Chalfonte-Haddon Hall 


Headquarters. The headquarters of the Convention will be the Chalfonte- 
Haddon Hall, where excellent accommodations have been assured, and a recep- 
tion suite will be provided for the Ladies' Committee. 

Reservations. Early in March room reservation cards were mailed to mem- 
bers of the Society. These cards should be returned as promptly as possible 
in order to be assured of satisfactory accommodations. 

Hotel Rates. Special rates have been guaranteed by the Chalfonte-Haddon 
Hall to SMPE delegates and their guests. These rates, European plan, will be 
as follows: 

Four Lower 

Floors Ocean View Ocean Front 

Room for one person $ 3.50 $ 4.00 $ 5.00 

Room for two persons 6.00 7.00 8.00 

Parlor Suite, for one 10 . 00 12 . 00 14 . 00 

Parlor Suite, for two 14 . 00 16 . 00 18 . 00 
(All bathrooms at Haddon Hall have hot and cold running fresh and salt 


If American plan rates are desired the hotel room clerk should be advised 
accordingly when registering. An additional charge of $3 per day per person 
will be added to the above-listed European rates for three daily meals, American 
plan. Members and guests registering at the hotel on the American plan will 
pay only $3 for the SMPE banquet scheduled at Haddon Hall on Wednesday 


evening, April 24th. If registered on the American plan, the clerk at registration 
headquarters should be advised accordingly when procuring your banquet 

Parking. Parking accommodations will be available to those who motor to 
the Convention at the Chalfonte-Haddon Hall garage, at the rate of 50fi for day 
parking or $1.25 for twenty-four hours. These rates include pick-up and delivery 
of car. 

Registration. The registration and information headquarters will be located 
at the entrance of the Viking Room on the ballroom floor where the technical and 
business sessions will be held. All members and guests attending the Convention 
are expected to register and receive their badges and identification cards required 
for admission to all the sessions of the Convention, as well as to several motion 
picture theaters in the vicinity of the Hotel. 

Technical Sessions 

The technical sessions of the Convention will be held in the Viking Room of 
the Hotel. The Papers Committee plans to have a very attractive program of 
papers and presentations. 

Luncheon and Banquet 

The usual informal get-together luncheon will be held in the Benjamin West 
Room of Haddon Hall on Monday, April 22nd, at 12:30 p.m. The forty-sixth 
Semi-Annual Banquet and Dance of the Society will occur on the evening of 
Wednesday, April 24th, in the Rutland Room of Haddon Hall an evening of 
dancing and entertainment for members and guests. 

Ladies' Program 

A specially attractive program for the ladies attending the Convention is being 
arranged by Mrs. O. F. Neu, Hostess, and the Ladies' Committee. A suite will 
be provided in the Hotel where the ladies will register and meet for the various 
events upon their program. 


At the time of registering, passes will be issued to the delegates of the Con- 
vention admitting them to the Apollo and Strand Theaters, by courtesy of 
Weilland and Lewis Theaters, Inc., and the Stanley and Virginia Theaters, 
courtesy of Warner Bros. Theaters. These theaters are in the vicinity of the 

Atlantic City's boardwalk along the beach offers a great variety of interests, 
including many attractive shops and places of entertainment. 


Convention Vice-President 



APRIL 22-25, 1940 

The Papers Committee submits for the consideration of the membership the follow- 
ing abstracts of papers to be presented at the Spring Convention. It is hoped that the 
publication of these abstracts will encourage attendance at the meeting and facilitate 
discussion. The papers presented at Conventions constitute the bulk of the material 
published in the Journal. The abstracts may therefore be used as convenient refer- 
ence until the papers are published. 

J. I. CRABTREE, Editorial Vice-P resident 

S. HARRIS, Chairman, Papers Committee 
L. A. AICHOLTZ, Chairman, West Coast Papers Committee 









The Control of Sound in Theaters and Preview Rooms; C. C. Potwin, Electri- 
cal Research Products, Inc., New York, N. Y. 

Acoustical science can now be applied to better advantage than ever before in 
the planning of modern motion picture theaters. A broader understanding of the 
purposes and principles of acoustical design and treatment is needed, however, 
to make this application universal. The Society is in a position to do much 
toward fulfilling this need. 

Greater attention should be given to the design and development of the basic 
theater structure. The shaping of surfaces for the control of sound reflections is 
effective and can be kept within a desirable architectural limit. Furthermore, 
such shaping can be made to function successfully if the basic design is developed 
to control reverberation. 

The all too prevalent idea that "the more acoustical material used, the better 
the results" should be discouraged. Acoustical materials can be used more 
efficiently if they are distributed asymmetrically with due regard to the geometry 
of the reflecting surfaces. In general, they should not be concentrated in large 
compact areas on single surfaces. This principle of treatment and its effect upon 
the acoustical characteristics of theaters is discussed. 



Instrumental measurements of the effect of surface parallelism upon the fre- 
quency reverberation characteristic of a rectangular room are shown. The results 
are of particular interest with respect to the acoustical treatment of preview rooms. 

Current Practices in Blooping Sound-Film; W. H. Offenhauser, Jr., Bernot- 
Maurer Corp., New York, N. Y. 

A review of our dimensional standards fails to indicate any attempt in the past 
to standardize sound-track bloops. While it is true that there is relatively little 
difficulty due to bloops at the present time, this condition appears to be due to 
the fact that each producing organization has more or less independently arrived 
at some rule-of-thumb solution to its particular problem rather than a result of 
any directed effort on the part of the industry as a whole. 

The volume of film affected is already very large and all indications seem to 
point to a substantial increase in the future. With this increase in prospect, it 
appears that an analysis of the subject is justified in order that standardization 
may be accomplished when, as, and if desirable. 

The criteria at the present time are almost entirely empirical; the common 
tests are (/) peak volume indicator and (2) listening. This has resulted in a wide 
variety of bloops in use ; a reduction in the number of sizes and types seems de- 
sirable in the interest of simplification. For single-track negative bloop punches 
this is especially important. 

In actual use, the length of the bloop punch varies from as small as 0.330 inch 
in one case to as large as 0.965 inch in another. A length of 0.500 inch may be 
considered to represent "average" practice. There is almost complete agreement 
on the following characteristics of bloop punches: (1) The punch should be 
sharp. (2) In the case of the triangle or trapezium types, there should be rounded 
corners at the base of the triangle. 

There is no similar agreement in the use of bloops for sound positives; in the 
case of release prints, this matter is not especially pressing since release negatives 
are usually re-recorded and have few if any splices. 

An Investigation of the Influence of the Negative and Positive Materials on 
Ground Noise; O. Sandvik and W. K. Grimwood, Kodak Research Laboratories, 
Rochester, N. Y. 

This paper deals with the effect of the negative sound-track upon the ground- 
noise of the print. Data are presented showing the influence of negative density 
and negative gamma on print ground noise for fine, medium, and course-grain 
negative emulsions. 

The Effects of Ultraviolet Light upon Variable-Density Recording; J. G. Frayne 
and V. Pagliarulo, Electrical Research Products, Inc., Hollywood, Calif. 

The effect of using ultraviolet filters upon the gamma of negative and positive 
development is discussed. The effect of ultraviolet light upon image quality is 
discussed, and a mathematical analysis is given explaining the existence of spurious 
side-images found in white-light recording on clear-base variable-density nega- 
tives. Low-end frequency rise, attributed to existence of these side-images, is 
eliminated by recording with ultraviolet. Reduction in wave-shape distortion as 


well as improvement in high-frequency response attributed more to use of ultra- 
violet in printing than in recording. Practically no gain in signal-to-noise ratio 
is found by using ultraviolet in either recording or printing. 

Photographic Tone Reproduction, Theory and Practice; Loyd A. Jones, Kodak 
Research Laboratories, Rochester, N. Y. 

For many years, in fact ever since the early beginnings of photography, many 
workers in the field have dealt with various phases, both theoretical and prac- 
tical, of the photographic tone reproduction problem. The word "tone" as used 
in this connection refers to the brightness and brightness differences existing in 
the original and in the photographic reproduction thereof. Hurter and Driffield, 
who were pioneers in the field of photographic sensitometry, gave some considera- 
tion to this problem, and since that time many contributions to the literature of 
the subject have been made by various contributors. The present paper aims 
to summarize the work which has been done in this field and to give an account 
of the present status. 

Some of the most recent work done in these laboratories in correlating theo- 
retical and practical aspects of tone reproduction will be discussed in some detail. 
This work has centered largely upon two subjects: the application of tone re- 
production theory to the development of a suitable criterion for expressing the 
effective camera speeds of negative materials used extensively in the field of 
amateur photography ; and the evaluation of the relative photographic quality 
of positives in terms of the amount of exposure given in making the negatives from 
which these positives were made. 

As a result of these studies, direct practical evidence has been obtained which 
verifies quite satisfactorily the theoretical conclusions previously reached to the 
effect that the gradient characteristics of both negative and positive sensitometric 
curves are of utmost importance in the determination of effective camera speeds 
and photographic positive quality as evaluated directly in terms of perceptual 

Tone Reproduction in Television; I. G. Maloff, RCA Manufacturing Co., Inc., 
Camden, N. J. 

The purpose of television is to produce moving pictures of original scenes in 
homes, auditoriums, and theaters. From the standpoint of the requirements of 
pictorial tone reproduction, television is closely related to motion pictures. 
However, the technic of tone reproduction in television is vastly different from 
that in motion pictures. The degree of perfection of pictorial tone reproduction 
of present-day television is, in some respects, not as high as that obtainable with 
35-mm motion pictures. On the other hand, the medium of television is the 
electrical signal, which is a great deal more flexible than photographic emulsions 
and permits effects unobtainable with the latter. 

The paper treats pictorial tone reproduction in television in detail. Means of 
obtaining desired range, contrast, perspective, and intensity, with adequate 
resolution, adequate illusion of motion and freedom from flicker, are discussed. 
Limitations and flexibility of pictorial tone reproduction in television are de- 
scribed in comparison with older methods of pictorial reproduction, and typical 


tone reproduction characteristics of the complete television system as well as its 
essential components are given. 

Direct 16-Mm Production; Lloyd Thompson, The Calvin Company, Kansas 
City, Mo. 

There are so many reasons why 16-mm film can and should be used that the in- 
dustrial and educational user is using more and more of it. The production of 
16-mm sound pictures by the direct method has been making progress. Today 
there are a number of companies using direct black-and-white and color sound 
productions in the 16-mm size. Many who are trying to use the method do not 
understand the proper technic or do not use the best commercial facilities available, 
which make the process slow in being generally accepted. 

Certain advantages and economies are effected by using the direct 16-mm pro- 
duction method which make it desirable for the non-user of sound-films to use this 
medium for the first time, and for others to use the film more effectively. Com- 
plete commercial 16-mm production and laboratory facilities are now available 
that equal those of the best 35-mm industrial producers. The problem of making 
wipes, fades, dissolves, and other tricks in the laboratory has been solved for 
'direct 16-mm production. Re-recording facilities for blending sound from several 
sources are available, making it possible to achieve truly professional results by 
the direct method. A few examples of direct 16-mm productions are given. 

Commercial Motion Picture Production with 16-Mm Equipment; John A. 
Maurer, The Berndt-Maurer Corp., New York, N. Y. 

Production of commercial sound motion pictures directly in the 16-mm size 
has increased rapidly during the past few years. Particularly in the production 
of those types of industrial films which are photographed in the field or factory 
rather than in the studio, the well known advantages of relative simplicity, port- 
ability, and freedom from fire risk in 16-mm equipment lead to economies that 
have frequently been decisive in making possible new applications of films. 

This paper surveys broadly the equipment, films, and services that are avail- 
able for 16-mm production, and presents a critical evaluation of the methods that 
are in use. 

Copies of 16-mm films are being produced at the present time by reversal dupli- 
cation from reversal originals, by making prints from reversal originals by means 
of an intermediate negative on fine-grain stock, by the direct negative-positive 
procedure, and by Kodachrome duplication. Prints produced by each of these 
processes will be demonstrated. 

Professional 16-Mm Recording Equipment; D. Canady, Canady Sound Appli- 
ance Co., Cleveland, Ohio. 

Details and description of 16-mm sound recording equipment for professional 
use is given, including: 

(1) A 16-mm recorder employing a high-quality optical system and glow-lamp. 

(2) Sound-track optical-reduction printer, permitting 16-mm variable-density 
sound-track being made from either 35-mm variable-density or variable-area re- 
corded track. 


(5) Noise-reduction equipment, for use in connection with glow-lamps or the 
new high-pressure quartz mercury lamp. 

Sixteen-Mm Equipment and Practice in Commercial Film Production; J. F. 

Clemenger and F. C. Wood, Jr., Sound Master, Inc., New York, N. Y. 

Today's commercial film is designed to accomplish a specific purpose and is 
therefore particularly directed to a specific audience. Prior to the introduction 
of the 16-mm sound projector, commercial sound-films could for the most part 
be shown only to theatrical entertainment audiences. 

The immediate acceptance and rapid growth in use of 16-mm sound projection 
equipment for the first time made it possible for the commercial film producer to 
select the audience most useful to him. 

At first practically all commercial 16-mm sound-films were made on 35-mm 
equipment and subsequently optically reduced to obtain 16-mm prints. It soon 
became obvious that it would be desirable to produce these films in the same 
medium in which they were to be shown. Among the advantages to be gained 
by such procedure were the absence of fire risk and consequent freedom from 
legal restrictions, the compactness and portability of equipment, lower raw-stock 
and print costs, and greater flexibility. 

The RCA Portable Television Pick-Up Equipment; G. L. Beers, RCA Manu- 
facturing Co., Camden, N. J.; O. H. Schade, RCA Radiotron Corp., Harrison, 
N. J. ; and R. E. Shelby, National Broadcasting Co., New York, N. Y. 

Spot news, athletic events, parades, etc., form an important source of television 
program material. Portable pick-up equipment suitable for televising such events 
has recently been developed. The equipment includes a small Iconoscope camera, 
camera auxiliary, camera control and synchronizing generator units, and a 325- 
megacycle relay transmitter and receiver. Most of the units are about the size 
of a large suitcase and weigh between 40 and 70 pounds. Each of the units is 
described and some of the practical applications of the equipment are indicated. 

Quality in Television Pictures; P. C. Goldmark and J. N. Dyer, Television 
Engineering Department, Columbia Broadcasting System, Inc., New York, N. Y. 

Present television standards specify certain factors that determine the ap- 
pearance of a television picture only to a limited extent. Other factors, however, 
such as contrast, gradation, brilliance, and the shape of the scanning spot are 
fully as important and are discussed in the paper. 

A photographic method of producing artificial pictures that permits varying 
several of these factors will be explained. Pictures will be shown that were ob- 
tained by this method and approach ideal quality within a given set of standards. 

A New Method of Synchronization for Television Systems; T. T. Goldsmith, 
R. L. Campbell, and S. W. Stanton, Allen B. DuMont Laboratories, Inc., Passaic, 

Line and frame scanning frequencies in an all-electronic television system need 
not be frozen to a standard giving limited definition performance if the synchroniz- 
ing system is arranged to allow flexible operation. Automatic operation of re- 


ceiver synchronizing circuits at variable line and frame frequencies is made pos- 
sible with the aid of a new type of synchronizing wave-form. Synchronizing 
standards which permit both flexible and automatic operation are discussed. 
Transmitter synchronizing apparatus for flexible synchronizing standards, re- 
ceiver circuits for both non-automatic and automatic synchronous operation are 
also discussed, and a "transition" type receiver for operation on both old and new 
type of synchronizing signals is briefly described. 

Advancement in Projection Practice; F. H. Richardson, Quigley Publishing 
Co., New York, N. Y. 

This paper briefly reviews projection practice from the beginning, pointing out 
the extremely poor conditions confronting projectionists in early days. By means 
of some twenty stereopticon slides the early projection equipments are illustrated 
and contrasted with those in use today. The work of some of the outstanding 
pioneers who had to do with early invention and improvements in projection 
equipments is described. 

Defects in Motion Picture Projection and Their Correction; I. Gordon, Akron, 

A statement is presented of the various kinds of damage inflicted upon screen 
images by oil on film. The paper enumerates the sources of this evil, the heavy 
loss the box-office can suffer as a result of them, the ill effect upon eyes of theater 
patrons, and suggests means for reducing the evil or possibly eradicating it. 

A Personal Safety Factor for Projection Practice; T. P. Hover, Lima, Ohio. 

The dangers inherent to the projection of nitrocellulose film are so obvious, and 
the accidents, when they occur, are so spectacular that practically no attention 
is given to other hazards in the projection room. This is to be expected in an in- 
dustry where practically no knowledge concerning the equipment and its opera- 
tion ever appears to the outside world. Only the joint cooperation of the manu- 
facturer of equipment, the sound supervisor, the theater manager, the pro- 
jectionists, and intelligent public safety officials can make the profession of pro- 
jecting motion pictures a safe one. Some of the observations of the author, who 
is closely associated with safety officials in the State of Ohio, are given in the 

Projection Supervision, Its Problems and Its Importance; Harry Rubin, Para- 
mount Theaters Service Corp., New York, N. Y. 

The importance of thorough and continuous supervision of projection and 
sound equipments in the theaters, some of the problems connected therewith in 
the construction and the maintenance of the theaters; a brief outline of a few of 
the many details that must be examined and precautions that must be observed 
in order that the motion picture entertainment may be presented under the most 
nearly perfect conditions, are described by a Projection Supervisor for a theater 
chain. Emphasis is placed upon the benefits to be derived through the close 
cooperation between the supervisor and the projection personnel of the individual 
theaters and several measures for accomplishing this result are cited. 


Products of Combustion of the Carbon Arc; A. C. Downes, National Carbon 
Co., Cleveland, Ohio. 

This paper is a review of work done in the laboratories of National Carbon 
Company, Inc., the College of Medicine of the University of Nebraska, the 
School of Public Health of Harvard University, and the Department of Health 
of the City of Detroit on the products of combustion from carbon arcs used in 
the motion picture industry. Analyses of the gases coming from various lamps 
show that, even in the stacks, the only gas occurring in toxic concentration is 
nitrogen dioxide. 

The biological effects of undiluted stack gas from simplified high-intensity arcs 
upon experimental animals were only those due to the nitrogen dioxide. 

The arc-ash fume when administered by intratracheal and subcutaneous 
routes in rabbits was found to be relatively inert. 

Determination of nitrogen dioxide concentrations in poorly ventilated projec- 
tion rooms failed to show any concentration more than about one-fifth that gen- 
erally considered as allowable for exposure of several hours duration, and there- 
fore there is little or no hazard in these projection rooms. 

Studies of ventilation under controlled conditions show that even with very 
low rates of both lamp house and room ventilation there is no danger of gases or 
fumes reaching concentrations which are toxic and that if sufficient ventilation 
is provided to produce comfortable working conditions there can not be any ap- 
preciable concentrations of nitrogen dioxide or arc-ash fumes in the booth. 

Rating of Motor-Generator Equipment Used for Direct Current Supply to 
Projection Arc Lamps; C. C. Dash, Hertner Electric Co., Cleveland, Ohio. 

The ratings of electrical equipment in general are based upon the heating and 
upon the performance. 

The projection room duty cycle with the alternate burning of two lamps and 
a single lamp puts a rather peculiar load upon generating equipment. This affects 
the temperature rating of the unit. The most important item in connection with 
the rating is the output characteristic. If designed for heating alone, a generator 
set would be unsatisfactory. 

The paper considers the output characteristics desirable for use with present- 
day arc lamps and their effect upon the design of the unit. Motor ratings will also 
be discussed. 

Records for the Projectionist; J. R. Prater, Palouse, Wash. 

Some portions of the data necessary to good projection room records may be 
kept to the best advantage on blank forms. Examples are shown of such blanks 
adapted to (1) an inventory of projection room supplies and spare parts, (2) 
data on vacuum tubes, (5) exciter lamps, (4) film inspection, and (5) a cue 
sheet. Noteworthy features are discussed, and suggestions given for adapting 
these forms to individual projection rooms. Projectionists will find it easier to 
keep good records on appropriate forms than it is to get by without them. 

Mathematical Expression of Developer Behavior; J. R. Alburger, RCA 
Manufacturing Company, Camden, N. J. 


Characteristics of developing agents have been unified in a mathematical ex- 
pression. The use of the analysis of developer behavior afforded by this ex- 
pression has been helpful in providing a guide toward improving a developer with 
respect to any given characteristic. 

Recording and Reproducing Square Waves; D. Canady, Canady Sound Ap- 
pliance Co., Cleveland, Ohio. 

A brief description of electrical equipment involved in the recording and 
reproduction of square waves is given. 

Direct-coupled amplification is used throughout as conventional amplifier cir- 
cuits are unsatisfactory when dealing with steep wave-fronts. Toe recording has 
been found satisfactory as picture requirements are not involved. 

Oscillograms and illustrations of mechanical wave-forms used in testing are 
shown and described. Records showing speech syllables passed through a con- 
ventional transformer-coupled system and direct-coupled amplifier are relatively 
striking and show the usual asymmetries encountered in a-c amplifiers when com- 
pared with direct-coupled systems. 

Motion Picture Theater Developments; by M. Rettinger, RCA Manufactur- 
ing Company, Hollywood, Calif. 

The first part of the paper is devoted to conveying basic requirements as well as 
recent developments in the design of motion picture theaters with balconies to 
provide satisfactory conditions for all the basic considerations of proper motion 
picture presentation. 

The second part is providing similar information pertaining to theaters with 
balconies. Separate sections are provided for the recommended dimensional and 
constructional features of balcony depth, soffit, and height; of the theater ceiling, 
sidewalls, and rear wall; and of the space above the balcony. 

Silent Variable Speed Treadmill; J. E. Robbins, Paramount Pictures, Inc., 
Hollywood, Calif. 

Treadmills are a definite necessity to the making of motion pictures for the 
purpose of obtaining intimate scenes of animated objects or persons working be- 
fore moving backgrounds. The evolution of this type of equipment dates back 
to the very beginning of the industry. Due to the fact that noise was of no con- 
sequence, these earlier machines were simply and crudely constructed. The type 
generally used employed the ordinary conveyor-chain principle, utilizing web 
belts running over series of rollers. Other developments include the revolving 
disk type, not entirely desirable due to the variation of surface speed in relation 
to the distance from the center of the circle; the gravity unit motivated by the 
persons or animals walking or running on them, etc., etc. Inasmuch as these were 
generally operated in front of sky backings or moving panoramas, speed ranges 
obtainable by gear boxes or belt pulley or chain sprocket changes were adequate. 
With the advent of sound and a more general use of the transparency or process 
background the need of smoother, more flexible, silent mills was recognized. The 
problem was carefully considered by the engineering department of Paramount 


Pictures, Inc., and the unit recently developed by them embodies all the pre- 
viously mentioned requisites and to date has operated satisfactorily under the 
most trying conditions. 

Construction details, speeds, degrees of silence, and other factors are covered in 
the paper. 

Optimum Load Impedance for Feedback Amplifiers; B. F. Miller, Warner 
Bros. First National Studios, Burbank, Calif. 

The apparent plate-resistance of vacuum-tubes employed in inverse feedback 
amplifier stages is shown to be a function of the degree of feedback employed. 
Equations for predicting the optimum value of amplifier load impedance for 
maximum undistorted power output are derived, and the necessity for properly 
building out the amplifier load circuit is demonstrated. A basic circuit, employ- 
ing a combination of two feedback elements is indicated, which permits securing 
the maximum undistorted power output from an amplifier stage while maintain- 
ing proper impedance relationships between amplifier and load circuits without 
the use of building out resistors. 

A Modern Studio Laboratory; G. M. Best and F. R. Gage, Warner Bros. First 
National Studios, Burbank, Calif. 

A description of the new laboratory erected by Warner Bros, at Burbank, 
Calif, in 1938. No general release work is required of this laboratory, and the 
generous space provided is devoted exclusively to the developing and printing of 
the dailies, storage and handling of the negative, and the latest in air-conditioning 
and dust-removing equipment. Advantage has been taken of the recent develop- 
ments in rust-resisting and acid-proof metals, especially in the construction of the 
developing machine tanks. The description includes the method of operation 
through the dailies, negative cutting, printing, chemical mixing, silver recovery, 
and other essential processes. 

Audience Noise as a Limitation to the Permissible Volume Range of Dialog 
in Sound Motion Pictures; W. A. Mueller, Warner Bros. First National Studios, 
Burbank, Calif. 

A series of noise measurements were made in theaters to determine the cause of 
low intelligibility of dialog recordings of wide volume range. Audience noise 
level was found to be a serious restriction, because it averages 8 db louder than 
film noise level and reduces the useful volume range by that amount. Audience 
noise is an extremely variable factor, as measurements made in the same theater 
showed it to be as low as the film noise in one instance and later to rise 14 db 
above this value. To secure good intelligibility, the volume range of the dialog 
must be compressed so that the softest-spoken words never are so low in level as 
to be seriously masked by audience noise. 

Color Theories and the Intersociety Color Council; H. P. Gage, Corning Glass 
Company, Corning, N. Y. 
Thanks to intensified study of color by scientists of the National Bureau of 


Standards, of the Agricultural Marketing Service of the U. S. Department of 
Agriculture, of the committees of the American Association of Railways, glass 
manufacturers, dye manufacturers, paint and ink manufacturers, the American 
Pharmaceutical Association, and photographic manufacturers, and the stimula- 
tion of the motion picture industry, the theories of color have been put in shape 
and tied together with extensive data on the color vision of many observers so 
that a workable engineering evaluation of colors, a scientific system of naming 
them, and a practical means of producing them to exact specification is now avail- 
able and is ripe for presentation not only to learned societies but to the general 

The phenomena and theory of the production of color in photographs both still 
and motion pictures have frequently been presented to this Society, and some 
phases will be rapidly reviewed in a demonstration of the spectral characteristics 
of color. 

Colored lights are subject to spectrophotometric measurement and by means 
of the I.C.I. (International Commission on Illumination) data can be inter- 
preted in terms of luminosity and the jc and y coordinates (or maps defining 

In these terms are being defined all standard Atlases of Color such as the 
Maertz & Paul Dictionary of Color, the Munsell Book of Color, and, it is hoped, 
the next standard set of colors of the Color Card Association used by all manu- 
facturers of clothing and other things in which standardization of manufacture in 
spite of rapidly changing styles is an economic necessity. 

The next edition of the National Formulary, sponsored by the American 
Pharmaceutical Association, will use this system of color names to describe the 
normal appearance of all drugs and chemicals. 

A shorthand method of describing the spectrophotometric analysis of color 
filters for theater spot and floodlights in the form of a seven-digit number has been 
devised for commercial specification of this material. 

These activities of numerous separate individuals and members of different 
technical societies have been coordinated and freely discussed by the delegates 
and individual members of the Intersociety Color Council so that all phases of the 
situation have been discussed. 

The Intersociety Color Council is made up of 74 delegates appointed by 11 
member societies, and by 67 individual members. It functions as a joint com- 
mittee on color of the member societies favored with the advice of the individual 
members. The Council issues News Letters in mimeograph form to its members. 
They contain information of progress in color work, notices of important color 
publications, the activities of the Color Council and notices of its planned meet- 
ings. It is not intended as a competing journal but with the minutes of the 
meetings serves as a basis for reports by the delegates to the member societies 
which can be published in their Journals. The Council sponsors meetings with 
the member societies on the subject of color. These papers are published in the 
journals of the societies. Such joint meetings have been held with the Optical 
Society of America, the Technical Association of the Pulp & Paper Industry 
(T.A.P.P.I.), the American Psychological Association, and a joint technical 
session on color will be held at the annual convention of the Illuminating Engi- 
neering Society this fall. 


The Cyclex System of Motion Picture Projection; C. S. Ashcraft, Ashcraft 
Mfg. Corp., Long Island City, N. Y. 

Cyclex is a new method of light projection, particularly adapted to the projec- 
tion of motion pictures. It is based on a method of coordinating light impulses, 
whereby alternating current may be used for the production of an electric arc 
and the light therefrom projected through a rotating shutter upon a screen, with a 
total absence of the periodic visual beat which has heretofore characterized 
alternating-current projection arcs. 

Simultaneously with the development of the non-pulsating alternating-current 
projection system, has been the development of a distinctly new type of alter- 
nating-current arc, wherein the characteristics of the arc itself have been used to 
the best advantage. Heretofore the method of operating alternating-current 
arcs has not been conducive to obtaining the highest efficiency and economy. 
The new Cyclex arc, however, has resulted in a light-source producing a far 
greater screen brilliancy together with a greatly reduced power input and conse- 
quent carbon consumption. 

The practical application of the combination necessitated the development of a 
radically new type of power conversion equipment, particularly adapted to the 
operation of the new arc. This equipment must have such characteristics that 
coordinated light impulses may be obtained with the maximum of electrical and 
mechanical efficiency, simplicity, and flexibility. 

The purpose of the present paper is (a) to present the basic theory of the system, 
(b) explain the characteristics of the Cyclex arc, together with the quality of the 
light produced, and (c) describe the apparatus employed for flexible frequency 
conversion and method of polyphase current transformation for the arc supply. 

Television Pick-Up of the Pasadena Rose Tournament Parade, January 1, 1940; 
H. R. Lubcke, Don Lee Broadcasting System, Los Angeles, Calif. 

The first television pick-up of the Pasadena Rose Tournament Parade was made 
on New Year's Day, 1940. This was accomplished with the "suitcase" type port- 
able television equipment and beam transmitter W6XD U of the Don Lee Broad- 
casting System. 

Two television cameras were used to give long-shot and close-up views of the 
floats; the cameras being arranged to give instantaneous switching of scene. 
The distance from Pasadena to the Don Lee Building, site of the home trans- 
mitter W6XAO, is nine miles and the line of sight was interrupted by two hills 
and buildings. Since the portable transmitter operates on a wavelength of less 
than one meter, much effort was therefore directed toward erecting high and 
efficient antennas at the transmitter and receiver. 

Diathermy machines, as used by the medical profession, were found to cause 
interference even on the beam transmitter frequency of 324 megacycles, indi- 
cating the need for proper shielding of such devices. 

The sound portion of the broadcast was sent over the nationwide Mutual Net- 
work. Camera work and aural description were adequately synchronized. Al- 
though rain fell during the parade and the morning was darkly overcast, written 
statements of reception from W6XAO lookers up to 15 miles away reported 
clear images, enabling them to read the names on the floats and discern other items 
of detail. 


Speed Up Your Lens System; W. C. Miller, Paramount Pictures, Inc., Holly- 
wood, Calif. 

The tendency of bare glass surfaces to reflect light has always presented a seri- 
ous problem in optics. New discoveries in the field of physics have resulted in 
methods of reducing these light reflections. One of these methods has proved 
practicable for general use in optical equipment. The reduction of reflections 
in treated systems has been so great that ghosts and flares are rarely encountered. 
The light no longer reflected by the glass surfaces is transmitted by the optical 
systems, increasing their efficiency. Camera lenses treated with the new process 
show an increase in speed of nearly a full stop. New applications of the process 
are being found almost daily. 

The Theory of Three-Color Reproduction in Motion Picture Photography; 
J. B. Engl, New York, N. Y. 

The theory of Three- Color reproduction of Hardy- Wurzburg gives the neces- 
sary conditions which have to be fulfilled in order to get a truthful reproduction 
of color. It is applied to the production of colored moving films. The possi- 
bilities of color corrections in film practice are discussed. Considerations of cost 
and of technical difficulties seem to lead to the conclusion that the most practical 
is the known method of color correction in the recording process by an artificial 
distortion of the color values of the subject. 

The theory allows an approximate computation of the necessary amounts of 
distortions. Truthful color rendering to a certain extent can be obtained in a 
predetermined way. The necessity of a systematic study of artificial color distor- 
tion is emphasized. 

A Precision Integrating- Sphere Densitometer; J. G. Frayne and G. R. Crane, 
Electrical Research Products, Inc., Hollywood, Calif. 

A densitometer employing an integrating sphere associated with a stable 
high gain amplifier is described. Densities up to 3.0 are read directly on a multiple 
scale logarithmic meter. Visual diffuse operation is attained by simulating 
average eye characteristic by inserting appropriate filters in optical path. 

Filtering Factors of the Magnetic Drive; R. O. Drew and E. W. Kellogg, RCA 
Manufacturing Co., Camden, N. J. 

A laboratory model of magnetic drive film phonograph was modified so that 
speed fluctuations of large and measurable magnitude and of frequencies ranging 
from Vt to 7 cycles could be introduced either into the sprocket rotation or the 
magnet rotation. The resulting speed variations at the drum were determined 
by means of a "wowmeter." The large ratios of flutter reduction indicated by 
these measurements show in part why the magnetic drive gives unsurpassed 
film motion. 



At a meeting held at the office of the Society on March 13th a collection of all 
the standards of the SMPE and the Academy Research Council, prepared for 
submittal to the ASA, was reviewed. Most of the projects had been approved 
by both organizations, but several of the standards, including that of the release- 
print sound-track dimensions, are still under consideration. 

In addition to all these items, which include all the standards published in the 
March, 1938, issue of the JOURNAL, arrangements were made for a revision of the 
glossary and a study of blooping patches. 

The next meeting of the Committee will be held some time in May. 


On March 20th a meeting of the Television Committee was held at the office of 
the Society, at which time the agenda for the present season was established and 
the Sub-Committees organized. A Sub-Committee was delegated to complete 
the work on the television glossary which had been started during the previous 
year and another Sub- Committee to finish the work on the television bibliograph. 

The scope of the Committee's work was enlarged to include reproducing 
characteristics of viewing devices, electrical and operating characteristics of 
pick-up devices, studio lighting and camera technic, characteristics of film- 
scanning apparatus, and the ultimate recommendations for standardization 
deriving from these studies. 

On March 29th the Committee held a meeting at the CBS Television Studio in 
New York for a demonstration of the test-films developed during the past few 
months for use on the iconoscope and dissector channels. 

The next meeting of the Committee is scheduled for May 3rd. 


The Sub-Committee on Projection Practice held two meetings, on February 
29th and March 14th, at the Paramount Building, New York. The activities of 
the Working Committees were discussed and plans were made for continuing the 
work on tolerances and on projection screen brightness. 

Arrangements have been made for a presentation on the subject of projection 
room fire regulations at the meeting of the Fire Marshals at Atlantic City, May 
4th, under the auspices of the National Fire Protection Association. The Work- 
ing Committee on the Projection Room Plans has begun a revision of the plans 
published in 1938 in order to bring them up to date, and the Power Survey Work- 
ing Committee is now preparing a report from the data derived from the thousand 
or more questionnaires sent out during the past season to the theaters of the 




At a meeting held at RCA Photophone Studios, New York, on March 13th, 
Dr. C. H. Cartwright of Massachusetts Institute of Technology, presented a lec- 
ture on the new reflection-reducing coated lenses. The lecture was followed by 
the projection of two identical films from two projectors adjusted identically, one 
projector employing a regular uncoated lens and the other projector employing a 
lens identical in all respects except that its surfaces were coated. 

Half the aperture of each machine was masked so that the two halves of the 
pictures from the two machines matched up on the screen. Definite comparison 
of the improvement in light transmission was thereby made possible. 

The meeting was very well attended, nearly 300 persons being present, and an 
interesting discussion followed the presentation. 


At a meeting held at the Western Society of Engineers in Chicago on February 
27th, Mr. W. C. Kalb of the National Carbon Company presented a talk on the 
subject of "Projection Light Then and Now." The paper described in detail 
the evolution of the electric arc as applied to the projection of motion pictures, 
beginning with the vertical arc using incandescent carbons, to the present super- 
high-intensity and the Suprex horizontal arcs utilizing the highly developed gase- 
ous cored-type of carbon. 

The meeting was very well attended and a lively discussion followed the pre- 


On March llth a dinner-meeting of the Section was held at the Hollywood 
Athletic Club, at which time two presentations were given : 

Mr. S. Charles Lee, theater architect, presented a talk on the engineering fea- 
tures involved in the design of motion picture theaters, and Mr. E. H. Marks of 
the National Theater Supply Company discussed the commercial aspects of 
screen illumination. 




Volume XXXIV May, 1940 


Progress in the Motion Picture Industry 455 

Chemical Analysis of Photographic Developers and Fixing 
Baths. . . .R. B. ATKINSON AND V. C. SHANER 485 

Motion Picture Theater Developments . . .M. RETTINGER 524 

New Motion Picture Apparatus 

The Resonoscope S. K. WOLF AND L. B. HOLMES 534 

Current Literature 539 

Highlights of the Atlantic City Convention 541 

Program of the Atlantic City Convention 545 

Society Announcements 548 





Board of Editors 
J. I. CRABTREE, Chairman 




Subscription to non-members, $8.00 per annum ; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscription or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or Hotel Pennsylvania, New York, N. Y. 
Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 

West Coast Office, Suite 226, Equitable Bldg., Hollywood, Calif. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1940, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is not 
responsible for statements made by authors. 


* President: E. A. WILLIFORD, 30 East 42nd St., New York, N. Y. 

* Past-President: S. K. WOLF, RKO Building, New York, N. Y. 

* Executive Vice-President: N. LEVINSON, Burbank, Calif. 

** Engineering Vice-President: D. E. HYNDMAN, 350 Madison Ave., New York, 
N. Y. 

* Editorial Vice-President: J. I. CRABTREE. Kodak Park, Rochester, N. Y. 

** Financial Vice-President: A. S. DICKINSON, 28 W. 44th St., New York, N. Y. 

* Convention Vice-President: W. C. KUNZMANN, Box 6087, Cleveland, Ohio. 

* Secretary: J. FRANK, JR., 356 W. 44th St., New York, N. Y. 

* Treasurer: R. O. STROCK, 35-11 35th St., Astoria, Long Island, N. Y. 


* M. C. BATSEL. Front and Market Sts., Camden, N. J. 

* J. A. DUBRAY. 1801 Larchmont Ave., Chicago, 111. 
** A. N. GOLDSMITH, 580 Fifth Ave., New York, N. Y. 
** H. GRIFFIN, 90 Gold St.. New York, N. Y. 

* P. J. LARSEN, 29 S. Munn Ave., East Orange, N. J. 

* L. L. RYDER, 5451 Marathon St., Hollywood, Calif. 

** A. C. HARDY, Massachusetts Institute of Technology, Cambridge, Mass. 

* H. G. TASKER. 14065 Valley Vista Blvd., Van Nuys, Calif. 

* Term expires December 31, 1940. 
** Term expires December 31. 1941. 


Summary. This report of the Progress Committee covers the year 1939. The 
advances in the cinematographic art are classified as follows: (I) Cinematography: 
(A) Professional, (B) Substandard; (II) Sound Recording; (III) Sound and Picture 
Reproduction; (I V) Television; ( V) Publications and New Books. 

In the field of cinematography there has been little to report this 
year in the way of new emulsions. The only item of outstanding 
interest is the introduction of coated camera lenses to reduce surface 
reflections. This improvement promises to have revolutionary ef- 
fects in this field. In the amateur field many high-speed emulsions 
have been made available for the first time to the amateur and there 
has been considerable improvement in projection equipment in both 
16-mm and 8-mm fields. The excellent pictorial and sound quality 
now obtainable by the 16-mm medium is making it possible to photo- 
graph and record directly on 16-mm film, thereby opening up a vast 
field for the 16-mm technic. 

In the field of sound recording one of the most significant advances 
of the year has been the use of fine-grain films in variable-density re- 
cording and printing. The use of these films brings about a marked 
reduction in ground-noise as well as improving the quality of the re- 
produced sound. 

In the field of television one of the most notable advances has been 
the introduction of mercury arcs to provide cool lighting on the tele- 
vision stage. 

Current hostilities in Europe and in the Orient have disturbed 
conditions there so that the Committee has no reports from its cor- 
respondents abroad. 

The Committee wishes to thank the following companies for sup- 
plying material and photographs for the report : Electrical Research 
Products, Inc.; Eastman Kodak Company; Ampro Corp.; General 

* Presented at the 1940 Spring Meeting at Atlantic City, N. J.; received 
April 15, 1940. 



Electric Co.; General Radio Corp.; RCA Victor Corp.; and 20th 
Century-Fox Studio. 

J. G. FRAYNE, Chairman 






(I) Cinematography 

(.4) Professional 

(1) Emulsions 

(2) Cameras and Accessories 

(3) Lenses 

(4) Studio Lighting 

(5) Color 
( 5) Substandard 

(1) Films 

(2) Cameras and Accessories 

(3) Projectors and Accessories 

II Sound Recording 

(1) General 

(2) Equipment 

(3) Recording Methods 

(III) Sound and Picture Reproduction 

(IV) Television 

(1) General 

(2) Lighting 

(V) Publications and New Books 

(A) Professional 

Last year the Progress Report called attention to the improvement 
of panchromatic emulsions as the outstanding advance of the year. 
This year there is nothing notable to report in this field, and this is in 
marked contrast to the advances made in emulsions for sound record- 
ing and printing purposes which are discussed later in this report. 

(1) Emulsions. An interesting paper by Schilling 1 discussed the 
various films now supplied by Agfa to the German market, including 
Superpan, Ultrarapid, Pankine H, and Finopan. Data on develop- 
ing properties, speed, color-sensitivity, and graininess are included. 


A faster, finer-grained panchromatic negative film known as 
Superior-^ (Type 126} was announced by Dupont in December. It 
was claimed to have about double the speed of Superior-7 and to re- 
tain the wide latitude and shadow detail rendering characteristics of 
the earlier product. 

The supply of raw materials for photographic manufacturing opera- 
tions in this country was not cut off at the outbreak of hostilities in 
Europe in September as it had been in 1914. During the quarter 
century that had elapsed, sources of supply of several materials, such 
as gelatin, optical glass, sensitizing dyes, and certain developing 
agents, were built up in the United States. 

Motion picture film and the properties of its support from the stand- 
point of safety were discussed by Sulzer, 2 who dealt with the control 
of fire hazard in the production, distribution, and use of cellulose 
nitrate film. His report also treated the characteristics of cellulose 
acetate film in relation to this problem. Reports from abroad indi- 
cated that a gradual change-over to acetate film stock for 35-mm film 
appeared to be under way in Germany and France. 3 

A paper (in Russian) by Pakshver and Mankash gave data for cal- 
culating the rate of evaporation of the solvent from cellulose ester 
dopes in film-coating machines. The simplest case is considered of 
the evaporation of acetone in still air at 50C from a solution contain- 
ing 27.5 per cent cellulose acetate without a plasticizer. 4 

A series of experiments by Charriou and Valette 6 on hypersensitiza- 
tian of emulsions showed that the effect of a given treatment tends to 
increase with increasing wavelength. The total sensitivity of an 
infrared-sensitive emulsion was found to reach a value seven times the 
initial sensitivity. Several interesting theoretical papers by Trivelli 
and Smith dealt with a number of problems relating to emulsions 
such as sensitometric and size-frequency characteristics, H&D speed 
versus average grain size, effect of grain size on finishing, development 
in relation to coating thickness, and resolving power and structure.' 

Of historical interest is the fact that the year 1939 marked the 
fiftieth anniversary of the introduction of roll film by George Eastman 
and the sale of the first film by him to Thomas Edison with which 
the latter and W. K. L. Dickson prepared a short length of the first 
motion pictures. These were viewed with the aid of a peep-show de- 
vice called the "Kinetoscope." The year also represented the 
hundredth anniversary of the first public announcement and demon- 
stration of the Daguerreotype process. 



(2) Cameras and Accessories. The 20th Century-Fox camera de- 
scribed in last year's report has been greatly improved and is now in 
constant use on production. The relation in size between this silent 
camera and the present blimp camera is illustrated in Fig. 1, the new 
camera being shown on the right. 

The new mobile camera crane devised by John Arnold and his co- 
workers at MGM Studios combines all the features in one assembly 
necessary to three-dimensional movements of a camera. By the addi- 

FIG. 1. Comparison of blimp camera (left) and 20th Century-Fox silent 

camera (right). 

tion of a "fifth wheel," the device may be turned in less than its own 
length. Mounted on this dolly, a central upright cylindrical post 
carries the counterbalanced arm that may be raised and lowered by a 
motor-driven jack or screw, permitting its use in very low-ceilinged 
sets. This arm carries the camera assembly at one end and a hand- 
wheel at the opposite end ; this handwheel varies the enclosed counter- 
balances that will balance a full 1000 pounds at the camera. This 
arm, with a radius of eight feet, swings through 360 degrees, or a 
complete circle. 



This is the only camera assembly that will enable the operator to 
shoot in every direction with equal facility within an area limited only 
by the length of the arm in transverse directions; this movement is 
unlimited in the direction of dolly travel. 

(3) Lenses. One of the outstanding advances this past year has 
been the introduction of "coated" lenses in the camera to reduce re- 
flection at the glass-air surfaces. The technics used by the two out- 
standing experimenters in this field, namely, C. H. Cartwright of 

FIG. 2. Fresnel lens "baby" spotlight. 

Massachusetts Institute of Technology and John Strong of California 
Institute of Technology, are essentially similar and consist of deposit- 
ing a transparent fluoride layer on the glass. For optimum perform- 
ance this layer should be l /t the wavelength of light. A set of coated 
lenses has been in use at Paramount, and it is claimed that an //2.3 
lens, when so treated, becomes equivalent to an //1. 6, meanwhile re- 
taining the depth of field, and the definition of the f/2.3. Another 
way of stating this would be that the loss of light in an Astro "Pan- 
Tachar" lens, which is normally about 41 per cent, becomes negligible 
when such a lens is treated to reduce surface reflections. While the 
deposited film on each lens surface is only four millionths of an inch 



in thickness, the lens may be handled and even washed without re- 
moving this film. Its iridescent magenta sheen, it is believed, will 
not interfere with color photography, as tests indicate no interference 
with either transmission or color correction. Undoubtedly, this im- 
provement in the optical world, when fully perfected, will prove the 
outstanding and most progressive step of the year, presaging future 
advantages of inestimable value to an industry so wholly dependent 
on "little pieces of glass." 

(4) Studio Lighting. The use of fluorescent lamps for motion 

picture photography is discussed 
in detail in a paper by I nman and 
Robinson, 7 given before the So- 
ciety at its 1939 Spring Meeting 
in Hollywood. These lamps have 
continued to be used in increas- 
ing numbers, particularly for 
close-ups in black and white. 
The daylight lamp works par- 
ticularly well in Technicolor 

New "Tulamp" auxiliaries 
operating one lamp at leading 
power factor, and the other at 
lagging power factor, minimize 
stroboscopic effects and allow 

alternating current operation 
FIG. 3. No. 5 photoflash lamp 

(Courtesy General Electric Co.). with greater efficiency. 

A small Fresnel lens spot 

("Baby Keglight") (Fig. 2), using 500 and 750-watt lamps, has come 
into general use, made possible by more sensitive film. The lamps 
are available in 50-hour life (MP type); 3380 K (with special filter) 
for Technicolor photography, and 3200K for use with Eastman's 
Type B Kodachrome film. 

A still smaller edition of this spot with Fresnel lens, using 150 and 
200-watt lamps, introduced by a number of manufacturers, has be- 
come quite popular, both in motion picture photography and amateur 
still and movie photography. 

Of particular interest to motion picture still photographers and, no 
doubt, of general interest to others in the industry are the new Day- 
light Blue Photoflash Lamp No. 21B and Photoflash Lamp No. 5. 


The first is a regular No. 21 foil-filled photoflash lamp with a blue 
lacquer to act as a light filter. The resulting light output gives very 
satisfactory colors with the regular type Kodachrome. The No. 5 
photoflash lamp introduced the past year by the Mazda lamp manu- 
facturers is unique because of its small size combined with a relatively 
great light output (Fig. 3). A large number of these lamps can be 
carried in one's coat pocket. When used with an efficient reflector, 
this little lamp will do the work of lamps several times its size. Its 
aluminum wire filling gives a fairly broad flash characteristic suitable 
for synchronization. 

(5) Color. In the field of color processes, Technicolor continued to 
dominate all other commercial production methods using 35-mm 
film. More pictures were made by this process than in any year 
heretofore. For example, nine Technicolor feature productions were 
in progress during the month of July. One of these, Gulliver's Travels, 
produced by Fleisher, represented the second feature-length cartoon 
ever made. The longest color motion picture ever produced, Gone 
with the Wind, was commented on very favorably for the beauty of its 
artistry and the quality of the color photography. 

The Telco process for 35-mm film was stated to utilize bipack nega- 
tives from which black-and-white prints are made on both sides of 
duplitized positive film. The positive image is swollen during proc- 
essing in proportion to the exposure gradation, and for that emulsion 
printed from the panchromatic negative, a red dye is introduced into 
the relief to fill the unswollen parts. After buffing and reswelling, a 
yellow dyed gelatin layer is applied. The process is repeated on the 
opposite side of the ortho negative where blue and green-dyed layers 
are used. 8 

Expansion of facilities was announced for two bipack processes, 
Cinemacolor and Magnacolor. A new plant with a stated capacity 
of one million feet of two-color prints each week was opened by Cine- 
color in Burbank, Calif., in March, 1939. Interest was developed in 
the monopack triple emulsion processes, Kodachrome and Agfacolor, 
although to date these have been used only for still photography. 

(B) Substandard 

Present trends in the substandard motion picture field make it pos- 
sible now to classify this phase of the industry into two groups : the 
16-mm or semiprofessional, and the strictly amateur 8-mm group. 

Continual improvements in film quality, combined with better and 


more powerful projection equipment, now make it possible to exhibit 
16-mm productions successfully before large audiences. This, to- 
gether with the generally lower production cost of 16-mm produc- 
tions, has been instrumental in supplanting 35-mm film in many 
fields in which it, before the advent of these improvements in 16-mm 
film, found exclusive use. Sixteen-mm equipment is, in general, 
smaller and more flexible than 35-mm equipment. During the last 
year manufacturers, awakening to the possibilities offered by this re- 
cording medium, have striven to develop 16-mm films, cameras, pro- 
jectors, and accessories which would provide the same facilities that 
heretofore have been available only to the professional user of 35-mm 
film. In this they have been eminently successful, and today users 
of 16-mm films have at their disposal equipment capable of delivering 
results favorably comparable to 35-mm productions. 

The necessity for making the original recordings on 35-mm film 
with subsequent reduction to the 16-mm width no longer exists. 
During the last year numerous complete advertising productions were 
recorded directly on 16-mm film. Some were recorded by the nega- 
tive-positive technic, while others, taking advantage of the improved 
quality of reversible films, made use of the somewhat longer method 
of making the original on reversible film and the prints on positive 
film from a dupe negative. 

Manufacturers of both 8-mm and 16-mm equipment have been 
forced, because of the unsettled conditions abroad, to confine their 
activities to the domestic market. This has severely curtailed pro- 
duction on much of the better projection equipment which, up to the 
time of the war abroad, was being exported in large quantities. These 
products were finding a ready market in professional use where sub- 
standard films particularly in the 16-mm width are used for enter- 
tainment purposes in theaters. 

The amateur and public in general seem to have become so accus- 
tomed to having sound with motion pictures that, with the exception 
of strictly home exhibitions, any motion picture film shown publicly 
is regarded as incomplete if it does not have at least a suitable musical 
accompaniment. For many otherwise satisfactory educational and 
entertaining pictures, a sound-track on the film is not feasible because 
of either the technical difficulties or the cost, or a combination of 
both. Not to be handicapped by these difficulties, these exhibitors 
have turned to the use of records. The libraries of the various record 
producers offer inexhaustible sound material of excellent quality 


which can be suitably selected to provide background music for almost 
any type of picture. This can not be considered as a substitute for 
sound-on-film as far as commentary is concerned ; however, when one 
considers that at least three-fourths of the time of the average trave- 
logue is consumed with music and that only one-fourth of the actual 
time is taken up by commentary, it can be seen that the use of records 
with suitably titled films can, with proper presentation, make a very 
satisfactory exhibition. The use of recorded music for picture ac- 
companiment has led to the development and introduction of numer- 
ous high-grade, compact, portable record-reproducing and record- 
cutting outfits during the past year. Most of this equipment is suf- 
ficiently sturdy and simple for amateur use. 

(7) Films. Several new films were introduced during 1939. In 
the 16-mm field Agfa Ansco introduced Triple-5 Superpan Reversible 
film. This new reversible film has an extremely fast, fine-grained 
panchromatic emulsion especially suitable for use under difficult light 

Kodak Super-^Y" Panchromatic safety film was introduced by the 
Eastman Kodak Company. This new film has the same speed as 
Cine Kodak Super-Sensitive Panchromatic, with slightly higher con- 
trast, less graininess, and better definition. Later a Cine Kodak 8 
Super-^f Panchromatic Safety Film was made available by the East- 
man Kodak Company. 

Agfa Panchromatic Reversible film was announced. This film is 
a medium-speed, fine-grain, completely color-sensitive film having a 
gradation especially suitable for outdoor use. 

Super- X Panchromatic "Reversal" 16-mm sound recording film was 
offered by the Eastman Kodak Company to provide a film of better 
sound-recording characteristics combined with better picture quality 
for simultaneous picture and sound recording purposes. 

The Eastman Kodak Company also brought out Safety Super-XX' 
Panchromatic Negative film for workers desiring a 16-mm negative 

Agfa Twin 8 Hypan Reversible film was brought out by Agfa Ansco 
and for the first time provides users of 8-mm equipment with a high- 
speed, fine-grained reversible film. 

The Gevaert Company of America made available a 9.5-mm film for 
owners of 9.5-mm equipment in the United States. This film is 
manufactured abroad, but it is processed by Gevaert laboratories in 
the United States. 



(2) Cameras and Accessories. There were no newly designed 16- 
mm motion picture cameras of American manufacture introduced in 
1939. However, a number of the European made cameras were 
introduced to the American market during this period. Outstanding 
among these was the Zeiss Movikon. This camera incorporates a 
coupled range-finder for the 50-mm//1.4 Zeiss Sonnar lens, and has 

FIG. 4. Lens extension for Magazine Cine Kodak (Courtesy 
Eastman Kodak Co.). 

interchangeable lenses all working in connection with the coupled 
range finder. The camera is equipped with variable speeds and hand- 
crank for either forward or reverse motion. 

There were several new cameras introduced abroad during the past 
year. The Siemens C-2 16-mm camera, made by Siemens-Halske, 
was introduced in England. This camera is said to be of the preci- 



sion magazine type and has four speeds of 8, 16, 24, and 64 frames 
per second. A feature of the camera is the automatic diaphragm 
change when changing from one camera speed to another. 

A new 8-mm camera was brought out by the Revere Camera Com- 
pany of Chicago. This new camera is known as the Revere Super 8. 
It is equipped with an //3.5 Wollensak lens and has a built-in view 
finder. It has variable speeds of 8, 16, 24, and 32 frames per second. 
This camera uses single-width 8-mm film. 

FIG. 5. Dual-operation Ampro projectors with tri-purpose amplifier. 

The Eastman Kodak Company introduced lens extension tubes for 
the magazine Cine Kodak (Fig. 4). This device is very convenient 
for obtaining magnified close-ups of such things as flowers. 

(5) Projectors and Accessories. Projection equipment has been 
generally improved during the past year and several new models 
have been offered. The Kodascope Models G and EE, introduced 
last year, have been improved and now provide features for accurately 
adjusting the relation of the lamp filament to the condensers, provid- 


ing better and more even illumination. In addition, these models are 
now provided with a hinged gate to facilitate threading of the film. 
In the 16-mm field, Ampro and Bell & Howell have introduced pro- 
jectors provided with sufficient illumination for auditorium and 
theater use. Ampro introduced two new models for dual operation 

FIG. 6. Kodascope 8 Model 70 (Courtesy Eastman 
Kodak Co.). 

with a tri-purpose amplifier capable of delivering 55 watts of undis- 
torted power (Fig. 5). The amplifier can be used automatically with 
either of the twin projectors as well as with a microphone when neces- 

The Filmaster was introduced by Bell & Howell. This projector 
follows the well established design of Filmo projectors. It is entirely 
gear-driven and is said to be exceptionally silent. It can be equipped 



with line-voltage lamps ranging from 300 to 700 watts. It is equipped 

with an //1. 6 projection lens with a newly designed optical system. 

In the 8-mm field, Eastman Kodak and Ampro have introduced new 

models of projectors. Eastman introduced the Kodascope 8 Model 

FIG. 7. RCA 16-mm sound projector. 

70 (Fig. 6). This newly designed projector has a highly corrected // 
1.6 projection lens, simplified threading, three-speed control switch, 
and will accommodate 300, 400, or 500-watt projection lamps. 

A new 16-mm sound motion picture projector (PG-170} was intro- 
duced by RCA. It was designed especially for use by schools, clubs, 
industrial organizations, etc. Among its many desirable features are : 
simplified threading, separate motor take-up, efficient optical sys- 



terns, and good accessibility of all major parts. The amplifier has 
an output of ten watts at five per cent distortion. The complete 
equipment is housed in two carrying cases which are easily portable 
(Fig- 7). 

Ampro introduced the Model A-8 8-mm projector. This projector 
is equipped with an //1. 6 projection lens and provides a 500- watt 
projection lamp. The projector operates on either a-c or d-c. It is 
provided with sufficient cooling for forward or reverse projection. 

FIG. 8. Black cap projection lamps (Courtesy 
General Electric Co.). 

The Revere Camera Corporation of Chicago introduced a newly de- 
signed 8-mm, all gear-driven projector. The projector is of an all- 
cast construction design provided with a 500-watt line-voltage pro- 
jection lamp and an //1. 6 projection lens. 

Manufactures of equipment abroad are still faced with mechanical 
complications by the necessity of supplying projection equipment for 
handling 8-mm, 9.5-mm, or 16-mm film. 

All the higher power lamps used for 8 and 16-mm motion picture 
projection, which include the T-12 bulb (750 and 1000-watt) and the 
T-10 bulb (400 and 500-watt) types are now being supplied with a 


black end coating (Fig. 8). This permits simplification of the louvers 
at the top of the lamp housing and provides better ventilation. 


(1) General. One of the most significant advances of the year in 
sound recording technics has been the adaptation of fine-grain films 
to variable-density sound recording. The interest in the use of these 
films for such work has been due to the possibility that they might 
offer an appreciable reduction of the background noise existing in 
present standard stocks. It was felt that a reduction in noise in the 
film itself was essential to accommodate fully the volume range ob- 
tainable in modern sound recording systems. The results of a test- 
ing program by several of the West Coast studios and by Electrical 
Research Products, Inc., were discussed at the 1939 Fall Meeting and 
published. 9 Hilliard 10 also reported on the status of this development 
as carried forward at the MGM Laboratories, and Daily 11 described 
the results obtained by Paramount. 

Besides tests on existing fine-grain emulsions, extensive investiga- 
tion was made on several experimental films, which were developed 
especially for this purpose. With certain films, the use of an im- 
proved type of high-pressure mercury arc offered a satisfactory illumi- 
nant. It was also found that the standard tungsten lamp might be 
used as an exposure source with certain optical systems and under 
certain development conditions. An improvement in signal-to-noise 
ratio of at least 6 db was reported for a fine-grain print from a fine- 
grain negative. Considerable improvement was also noted in overall 
quality when fine-grain film was used for the original negative, re- 
recorded print and re-recorded negative, with the final print being 
made on standard positive film. Improved image definition is also 
claimed to result, probably due to reduced flare in the emulsion. The 
results of this development are sufficiently encouraging, according to 
the Committee Report 9 that in spite of attendant difficulties, the 
introduction of the technic on a wide scale into the motion picture 
industry appears to be inevitable. 

There is considerable activity in the studios in adapting fine-grain 
films to their recording programs. Paramount Pictures, Inc., have 
employed fine-grain film for all four steps: original, negative, dub- 
bing prints, release negative, and release prints. The use of fine-grain 
film for release negative at normal gamma complementary to standard 
release print gamma was made possible by the design of an appropri- 


ate developer providing satisfactory density at moderate developing 
time. Metro-Goldwyn-Mayer Studios have adopted fine-grain film 
for original negative, exposed by incandescent light (at above normal 
gamma), and dubbing prints (at complementary gamma). By virtue 
of the higher negative gamma, satisfactory density is obtained in a 
standard developer. To obtain sufficient exposure on fine-grain 
sound negatives, the mercury lamp has been adopted by several stu- 
dios, including Paramount, Metro-Goldwyn-Mayer, Samuel Goldwyn, 

FIG. 9. RCA unidirectional microphone. 

and Universal. A forced-draft cooling system has been developed by 
Paramount, allowing the lamp to be operated at several times the 
normal rated output. All these above-mentioned studios are regu- 
larly using fine-grain film for dubbing prints. 

Interest in dynamic testing methods applied to recording systems 
still continues. Use is made of the intermodulation, or two-fre- 
quency method and square- wave generators, or sound-tracks having 
this type of wave-form recorded upon it. Some measure of transient 
effects is given by the last two methods, but to date no adequate cor- 
relation between measurement and listening has been established. A 



large amount of work is required before satisfactory conclusions can 
be drawn from these methods. 

(2) Equipment. An outstanding technical demonstration to the 
industry was made by Bell Laboratories in the showing of the Voco- 
der. 12 The device synthesizes speech or music into its components 
and then remakes it to form the original. Circuit elements are adjust- 

FIG. 10. RCA experimental double-width push-pull variable-area 

able so that certain alterations can be made in the coded material be- 
fore it is recombined. This allows changes in pitch fundamental and 
inflection and permits vibrato effects to be added, completely altering 
the original characteristic of the material. Possible applications to 
sound recording are: studies of fundamental nature of speech and 
the way in which its composition is altered by changes in effort; 
alteration of the characteristics of original speech for improvement of 
intelligibility; or creation of new types of voices for special or cartoon 


effects. Undoubtedly other uses will suggest themselves as further 
experience is gained with the equipment. 

RCA introduced a unidirectional microphone having relatively high 
sensitivity and a wide frequency range. This unit was developed 
especially for sound-film recording and is designed for suspension 
mounting from a boom. The microphone has directional characteris- 
tics of the cardoid type and affords a wide angle of pick-up with uni- 
form frequency response. Undesirable sounds such as camera noise 

FIG. 11. Precision integrating sphere densitometer (Courtesy Electrical Re- 
search Products Inc.). 

and backstage reflections may be attenuated by the proper use of the 
directional characteristics. The microphone also tends to minimize 
the effects due to unfavorable studio acoustic conditions (Fig. 9). 

Bell Laboratories have added further versatility to the cardoid 
microphone by providing switching facilities which give three addi- 
tional directional patterns falling between the cardoid and bidirec- 
tional patterns. These patterns may offer improvement for difficult 
pick-up conditions. 

A push-pull shutter was developed for use with the RCA variable- 
density recording systems. The shutter has two penumbra vanes 


which move apart as the noise-reduction current increases. A fixed 
vane behind each of these acts as an optical limiter. The actuating 
motor is of the magnetic type and is similar to the motors employed 
in variable-area shutters. The standard and push-pull variable- 
density shutters are interchangeable in the recording optical system 
without alteration. One of the new shutters has been in use at the 
20th Century-Fox Studios for several months. 

An experimental recording optical system for making double- width 
class A push-pull sound-tracks was developed by RCA. Adequate 
exposure for ultraviolet recording is obtained by the use of a high- 
pressure mercury vapor lamp and a special power supply unit. A 
corrected spherical-cylindrical objective lens images the slit on the film 
at 7.5 to 1 reduction in one plane and 3.5 to 1 reduction in the other 

FIG. 12. General Radio square-wave generator. 

plane. The optical system base, the noise-reduction shutter, and the 
galvonometer remain unchanged. One complete system was built 
and is undergoing a series of field tests in Hollywood (Fig. 10). 

ERPI has developed a 200-mil push pull variable-width modulator 
which is capable of exposing standard sound emulsions for ultraviolet 
recording or fine-grain emulsions for white-light recording with an 
ordinary tungsten filament lamp. This modulator has been used to 
record original scorings in some of the Hollywood Studios during the 
past year. 

ERPI has announced a precision integrating sphere densitometer 
(Fig. 11). The device affords a rapid and accurate means of deter- 
mining densities by electrical means. Briefly, a light-beam is inter- 
rupted by means of a chopper and the modulated light, after passing 
through the film whose density is to be measured, actuates a photocell 
which is mounted within a six-inch integrating sphere connected to a 


stable amplifier and metering system. Densities are read directly 
upon a large-scale meter. 

General Radio have provided a square- wave generator 13 for trans- 
mission testing purposes. It is used in determining the frequency re- 
sponse, particularly under transient conditions, of amplifiers and other 
networks. The type 769- A square- wave generator shown in Fig. 12 
is a device for converting a sinusoidal timing signal into a square- wave 
signal. Squaring is accomplished by amplifying a sinusoidal signal 
and clipping both positive and negative peaks. This partially 
squared signal is reamplified and clipped a second time. The final 
signal is then fed through a phase inverter and amplified in a balanced 

ERPI has developed an intermodulation meter 14 which is applicable 
to the two-frequency method of transmission testing. It has already 
found many applications in sound-picture work for determining opti- 
mum conditions for film processing. 

The class B push-pull variable-area system continues to gain favor 
as a means of making original recordings. The Republic Studios 
have converted their recorders from class A push-pull to class B push- 
pull. After several months of production experience with this sys- 
tem, the studio has reported an appreciable reduction in noise, an 
improvement in quality, and a simplification of the recording opera- 
tions. 15 Extensive tests have also been made with the class B system 
at Warner Bros. Studios and at the Walt Disney Studios. 

Improvements in the performance of RCA noise-reduction ampli- 
fiers have been made through the use of exponential tubes. By in- 
creasing the margin for low-level sounds, it is possible to use smaller 
bias lines, thus affecting a substantial reduction in ground-noise dur- 
ing quiet passages. Initial clipping is also reduced by this new de- 
velopment. Exponential noise-reduction systems are in use at RKO 
Studios, Walt Disney Studios, and the Republic Studios. 

A new high-fidelity cutting head (MI-4887) was introduced by 
RCA. It is similar in design to the MI-4885 cutting head except that 
the frequency range has been extended from 7500 cycles per second to 
10,000 cycles per second. The new head may be used to cut either 
wax or lacquer disks (Fig. 13). 

(S) Recording Methods. RCA have described methods for secur- 
ing amplitude control in variable-density recording by optical means. " 
Further use of class B push-pull variable-density recording has also 
been related. 16 



Factors entering into recording and reproducing system characteris- 
tics which relate the effort employed by a speaker on the set to a 
proper reproduction of the same speech in the theater have been in- 
vestigated. 17 

Two interesting methods of producing artificial reverberation have 
been described. In the first, 18 the material to which reverberation 
is to be added is recorded upon a steel tape. A number of reproducing 
heads are separated by appropriate time intervals along the tape, so 
that when reproduced, multiple sources are obtained. The outputs 
are separately attenuated, thus simulating reverberation. The sec- 
ond method 19 employs an electro- 
optical system. Here the ma- 
terial is recorded on phosphores- 
cent material coated on the rim 
of a revolving turntable. The 
images are transitory, and by 
scanning the images at appro- 
priate intervals, a time delay and 
decay in volume output from 
these image sources are obtained. 

At the ERPI West Coast 
Laboratory an outdoor micro- 
phone test set-up has been es- 
tablished (Fig. 14). The equip- 
ment is similar to that employed 
by Western Electric and the Bell 
Telephone Laboratories, thus 
making possible comparable measurements by the manufacturer, 
designer, and user of microphones. In addition to use in checking 
the performance characteristics of microphones, correlation between 
listening tests and acoustic measurements has established criteria 
which indicate a microphone's acceptability for sound recording from 
its measured acoustical response. 


During the year 1939 there has been little to report in the way of 
new sound picture projection equipment. However, installation of 
systems previously reported continued at a good pace throughout the 
country. Several pieces of auxiliary equipment were developed 
through the year. 

FIG. 13. 

RCA 10,000-cycle recorder 


A new preview attachment (MI- 107 5) was introduced by RCA. 
It consists of a single lower magazine for a 35-mm theater type pro- 
jector. The new magazine accommodates three 1000-ft. reels for 
use when the sound and picture are simultaneously reproduced from 

FIG. 14. ERPI West Coast outdoor microphone test set-up. 

separate films. For normal projection of a composite print, space is 
provided for a 2000-ft. reel (Fig. 15). 

The year 1939 has again demonstrated the outstanding values of 
the Simplex E-7 projector mechanism which was first marketed during 
1 938. The introduction of dual shutters operating on the same shaft 
has made practical, through optical inversion, a total reduction in 
shutter blade area of 40 degrees, thereby appreciably increasing the 



picture brilliancy with a given light-source. Other important fea- 
tures include: a removable fire-trap; a "one-shot" pressure oiling 
system completely filtered to prevent foreign matter from getting to 
the bearings through the lubricating system ; automatic locking of the 
gate in either the open or closed positions; convenient means for ad- 

FIG. 15. RCA preview attachment. 

justing the film tension; and improved intermittent movement and 
sprocket, hardened and ground to such precision as greatly to reduce 
the movement of the picture at the aperture. 

The Simplex Four-Star Sound System, introduced during 1938, 
continued to be very popular with exhibitors during 1939. Most 
prominent among its design features are the ease of servicing, facili- 
ties for accurate adjustments, and the stability of its operating charac- 



teristics. Other innovations such as the extensive use of feedback 
and the reduction in the number of transformer-coupled circuits, re- 
sulted in a pronounced improvement in the matter of both amplitude 
and phase distortion. Pioneered in this country was the use of 
permanent magnet loud speakers, a feature which further emphasizes 
the Simplex theme of simplicity and stability. 

During 1939 International Projector Corporation introduced a 
completely new projection and sound equipment for the smaller type 

of theater. The Simplex SI pro- 
jector, together with a new low- 
cost pedestal of the rectangular 
box type and two lamp houses 
known as the Simplex Low and 
Simplex High, provide an attrac- 
tive line of quality apparatus for a 
very large market. Simultane- 
ously with this came the Type 
Four-Star Sound System which 
was designed to provide the same 
grade of high-quality reproduction 
for the smaller theaters as has been 
available in the past only to the 
larger and de luxe houses. In 
carrying out their design premises, 
this system furnishes exactly the 
same sound head as is used in 
all of the other Simplex Four-Star 
Sound Systems. In addition, the 
identical type of permanent-mag- 
net loud speaker units are used 
and the circuit features are fundamentally the same. The cost 
reduction of this system was primarily obtained through the use of 
a-c for energizing the exciter lamp, one common volume control 
amplifier instead of individual ones for each machine, cabinet and 
mounting arrangements, and the use of smaller multicellular high- 
frequency horns made possible by an 800-cycle dividing network. 

A new inexpensive brightness meter (Fig. 16) known as the Luc- 
kiesh-Holladay meter has recently been made commercially available 
by the General Electric Company. This covers a range of from Vioo 
of a foot-lambert to 75,000 foot-lamberts, and employs the familiar 

Fig. 16. Luckiesh-Holladay 
Brightness Meter (Courtesy Gen- 
eral Electric Co.). 


concentric spot field, part of the field being the object whose bright- 
ness is measured, and part by a small battery-operated lamp whose 
brightness is adjusted by means of a photocell-operated light-meter. 
It has already proved to be a useful instrument for motion picture 
screen brightness measurements as well as for general brightness sur- 
veys in the theater. 

Projector Lenses. The application of a non-reflecting coating sur- 
face to projection lenses similar to that discussed previously for camera 
lenses has been discussed by Prof. C. H. Cartwright of Massachusetts 
Institute of Technology, and was demonstrated before the Atlantic 
Coast Section of the Society. The demonstration showed the differ- 
ence between the new film-treated lens and a standard projection 
lens. Two matched films of the same subject were projected simul- 
taneously upon a single screen by two projectors, one equipped with 
a treated lens and the other with an untreated lens. Half of the film 
was masked in each projector and the halves matched on the screen, 
allowing visual comparison of the light-transmission of the two lenses. 

Miscellaneous. The past year has witnessed an increasing use of 
fluorescent materials, activated by ultraviolet light, both in motion 
picture photography and in the theater. Probably much of this has 
been due to the recent availability of simple, powerful, ultraviolet 
sources as well as improvements in the fluorescent materials them- 
selves. Lamps used for this purpose are of the electric discharge type 
of 100 and 25-watt rating. The filtering material that removes the 
visible radiation may be incorporated in the lamp bulb glass or in a 
roundel covering the front of the projector. One of the latest stunts 
is to impregnate the aisle carpets used in motion picture theaters with 
fluorescent material. Under the excitation from invisible ultraviolet 
projectors, the carpet glows with sufficient brightness to guide the 
patron down the darkened aisle. 


(1) General. Regular television service to the metropolitan New 
York area from a transmitter on the Empire State Building was in- 
augurated on an experimental basis on April 30, 1939, with fanfare 
coincident with the opening of the New York World's Fair. At about 
the same time television receivers by a number of manufacturers were 
offered for sale in the New York area. Transmissions have averaged 
approximately eleven hours per week of entertainment features plus 
a somewhat greater number of hours of test-pattern signals. Much 


work has been done on the development of program production tech- 
nic and a systematic study of audience reaction to the individual 
programs. Transmissions have been in accordance with the stand- 
ards recommended by the Radio Manufacturers Association. Tech- 
nical performance was considered satisfactory. The antenna is of a 
type having a uniform impedance over a band greater than one tele- 
vision channel as a result of its unusual configuration. The vestigial 

FIG. 17. Three AH-6 water-cooled lamps in single re- 
flector (Courtesy General Electric Co.). 

sideband signal is obtained by a special filter network at the output 
of the transmitter. A second New York station was nearly com- 
pleted. Experimental transmissions were also available in the Los 
Angeles, Chicago, Philadelphia, and Schenectady areas. Television 
demonstrations were features of both the New York and San Francisco 
Fairs and attracted large crowds at other places. 

Announcement was made of the development of a new type of pick- 
up tube which uses a low- velocity electron scanning-beam with resul- 
tant great improvement in image quality through the elimination of 
extraneous signal components. 



Programs picked up at points remote from the studio have become 
so popular that a new type portable equipment in suitcase form has 
been developed, manufactured, and put into service in the New York 
and Los Angeles areas. Some of the cameras for this .service use a 
small Iconoscope to make possible a small and light unit. A number 
of simplified television studio systems were built and used at fairs and 
exhibitions. It was found practicable to use short lengths of selected 
regular telephone circuits for the transmission from the remote point 
to the transmitter. 

FIG. 18. Scene being televised with cool light from 
water-cooled quartz mercury arc lamp, in G-E Television 
Studio, Schenectady, N. Y. 

Two reports dealing with problems in television broadcasting were 
issued by the television committee of the Federal Communications 

A beginning was made in industry cooperation toward reduction of 
various types of interference to television and other ultra-high-fre- 
quency services. 

Research continued on apparatus for large screen television and a 
number of showings were made to indicate status and progress. 

(2) Lighting. The lighting of the people being televised at the 
RCA exhibit at the World's Fair was accomplished by reflector in- 
candescent lamps of 300-watt rating, described in detail in Mr. Eddy's 
paper 20 given before the 1939 Spring Meeting of the Society. 


Three Type H-6 water-cooled lamps 21 were used to light the tele- 
vision studio in the General Electric exhibit (Fig. 17). Two incandes- 
cent spots, with heat-absorbing glass niters, were added to improve 
the appearance. 

Early in June, General Electric's experimental television studio at 
Schenectady began operation. For lighting the area in which the 
action occurs, twelve Type H-6 water-cooled lamps, arranged three 
to a reflector, are being used (Fig. 18). The purpose of putting three 
lamps in a single reflector is to minimize the effect of the cyclic varia- 
tion of the light, each lamp being placed on the leg of a three-phase 
circuit. This installation is capable of providing 1000 to 1200 foot- 
candles over an area 12X12 feet. 


A new publication of interest to motion picture technicians made 
its debut in June, 1939, under the name, Photo Technique, published 
by McGraw-Hill Book Company. A German technical journal 
also made its initial appearance during 1939, its title being Zeitschrift 
fur Angewandte Photographic and its publisher, S. Hirzel of Leipzig. 

The following books of noteworthy interest were published since the 
last report of the Committee in April, 1939: 

(1) Handbook of Photography; K. Henney and B. Dudley (McGraw-Hill 
Book Co., New York). 

(2) The Photographic Process; J. E. Mack and M. J. Martin (McGraw-Hill 
Book Co., New York). 

(3) Colour Cinematography; A. Klein (Second Edition) (Chapman and Hall, 

(4) Colour in Theory and Practice; H. D. Murray and D. A. Spencer, Vol. 1 
(Chapman and Hall, London). 

(5) Sound Motion Pictures, Recording and Reproducing; J. R. Cameron 
(Cameron Publishing Co., Woodmont, Conn.). 

(6) The Amplification and Distribution of Sound; A. E. Greenless (Chapman 
and Hall, London). 

(7) Tonfilm-Anlagen und ihre Bahandlung (Sound-Films and Their Treat- 
ment); F. Kleffel (W. Knappe, Halle, Germany). 

(8) Applied Acoustics; H. F. Olson and F. Massa, 2nd Edition (P. Blakiston's 
Son and Co., Philadelphia). 

(9) Motion Pictures and Radio (of educational interest) ; E. Lane (McGraw- 
Hill Book Co., New York). 

(10) Make Your Own Movies; A. Gale and K. Pressels ( Coward- McCann Co., 
New York). 

(11) Facts and Figures for the Amateur Cinematographer ; G. P. Kendall 
(Newnes, London). 


(12) Cine-Photography for Amateurs; J. H. Reyner, 3rd Edition (Chapman 
and Hall, London). 

(13) Professional Quality on Amateur Reversal Film; P. C. Smethurst 
(Link House Publication, London). 

(14) Photography by Infrared; W. Clark (Chapman and Hall, London). 

(15) Electron Optics in Television; I. G. Maloff and D. W. Epstein (McGraw- 
Hill Book Co., New York). 

(16) Television; V. Zworykin and G. A. Morton (J. Wiley and Sons, New 

(17) The Cinema as a Graphic Art; V. Nilsen (Trans, by S. Gary) (Newnes, 

(18) The History of Photography; E. Stenger (Trans, with footnotes by E. 
Epstean (Mack Printing Co., Easton, Penna.). 

Yearbooks were issued by the following publishers: 

Quigley Publishing Co., New York. 

Film Daily, New York. 

Kinematograph Publications, Ltd., London. 

Photo-Kino Verlag, Berlin. 

M. Hess, Berlin-Schonberg. 

Abridgments and compilations were issued as follows: 

Abridged Scientific Publications of the Kodak Research Laboratories, Vol. 20 
(1938) (Eastman Kodak Company, Rochester, New York). 

Jahresbericht der Photographic, Kinematographie und Reproduktionstechnik 
fur das Jahr 1937 (Review of Photography, Cinematography and Reproduction 
Processes for the Year 1937); H. Frieser and H. Staude (Akademische Verlags 
M. B. H. Leipzig). 

Veroffentlichungen des wissenschaftlichen Zentral-Laboratoriums der Photo- 
graphischen Abteilung Agfa (Publications of the Agfa Central Photographic 
Research Laboratories), 6 (S. Hirzel, Leipzig). 


1 Veroff. Wiss., Zentral Labs. Phot. Able. Agfa, 6 (1939), p. 65. 
1 Internal. Proj., 14 (April, 1939), p. 10. 

3 Phot. Ind., 37 (March 22, 1939), p. 403; also Filmtechnik, 14 (Oct., 1938), p. 

4 Kino. Photo. Chem. Ind. (U.S.S.R.), 21 (1939), No. 5. 

6 Publ. Set. Tech. Ministere de I'Air, Bull. Services Tech. No. 86 (1939). 

6 Phot. J., 79 (May, 1939), p. 330; 79 (July, 1939), p. 463; 79 (Nov., 1939), p. 
609; 79 (Dec., 1939), p. 630; 80 (Jan., 1940), p. 12. 

7 INMAN, G. E., AND ROBINSON, W. H.: "The Fluorescent Lamp and Its 
Application to Motion Picture Studio Lighting," /. Soc. Mot. Pict. Eng., XXXIII 
(Sept., 1939), p. 326* 

8 Kinemat. Weekly, 265 (March 30, 1939), p. 37; also Internal. Phot., 10 (Jan., 
1939), p. 10. 


/. Soc. Mot. Pict. Engs., XXXIV Qan., 1940), p. 3; also Internal. Phot. 11 
(Oct., 1939), p. 5. 

Amer. Cinemat., 20 (Dec., 1939), p. 535. 

/. Soc. Mot. Pict. Engs., XXXI V (Jan., 1940), p. 12. 

11 DUDLEY, H.: "Remaking Speech," J. Acous. Soc. Amer., (Oct., 1939), p. 

11 ARGUIMBA, L. B.: "Network Testing with Square Waves," General Radio 
Experimenter, XTV (Dec., 1939). 

14 FRAYNE, J. G., AND SCOVILLE, R. R.: "Analysis and Measurement of Dis- 
tortion in Variable-Density Recording," J. Soc. Mot. Pict. Eng., XXXII (June, 
1939), p. 648. 

u BLOOMBERG, D. J., AND LOOTENS, C. L.: "Class B Push-Pull Recording for 
Original Negatives," J. Soc. Mot. Pict. Eng., XXXIII (Dec., 1939), p. 664. 

18 DIMMICH, G. L.: "Optical Control of Wave-Shape and Amplitude Charac- 
teristics in Variable-Density Recording," /. Soc. Mot. Pict. Eng., XXXIII (Dec., 
1939), p. 650. 

17 LOYE, D. P., AND MORGAN, K. F.: "Sound Picture Recording and Repro- 
ducing Characteristics," J. Soc. Mot. Pict. Eng., XXXII (June, 1939), p. 631. 

u WOLF, S. K.: "Artificially Controlled Reverberation," J. Soc. Mot. Pict. 
Eng., XXXH (April, 1939), p. 390. 

19 GOLDMARK, P. C., AND HENRI CKS, P. S. : "Synthetic Reverberation," 
J. Soc. Mot. Pict. Eng. XXXIII (Dec., 1939), p. 635. 

EDDY, W. C.: "Television Lighting," /. Soc. Mot. Pict. Eng., XXXIH (July, 
1939), p. 41. 

11 NOEL, E. B., AND FARNHAM, R. E.: "A Water-Cooled Mercury Arc," J. 
Soc. Mot. Pict. Eng., XXXI (Sept., 1938), p. 221. 



Summary. Procedures for the qualitative and quantitative determination of the 
usual constituents of photographic developers and fixing baths are given. Volumetric 
methods are employed in general. 

The accuracy of the procedures and the significance of analytical findings are dis- 
cussed in relation to the functions of the various constituents of developers and fixing 
baths, both when fresh and after use. Suggestions are given as to the manner of appli- 
cation of analytical tests to typical cases of troubles arising from defective processing 

The selection of reagents, the use of special procedures and equipment, and the 
computation of results are treated in detail. 


Photography has developed from infancy as an art rather than a 
science, and it is only with the demands for uniformity and economy 
which come with large-scale commercial applications that the science 
of photography has been given attention. Photographic materials go 
through three important stages in their use, namely, manufacture, 
exposure in the camera or other mechanism, and processing. Scien- 
tific methods have been introduced widely into the first two phases of 
photography, and in recent years sensitometric control has been 
applied extensively in certain types of photographic processing, 
especially in handling motion picture film. The final stage to which 
scientific control has come is processing. One reason for this tardiness 
is the fact that the process is complicated, and adequate tools have 
only recently been developed. Primary among such tools are ana- 
lytical methods for the study of the solutions themselves. Such 
methods have been worked out on a practical basis with special refer- 
ence to motion picture solutions, and the methods here described have 
been found useful in handling problems of the motion picture film 

* Presented at the 1940 Spring Meeting at Atlantic City, N. J. 
** Eastman Kodak Co., Hollywood, Calif. 


486 R. B. ATKINSON AND V. C. SHANER [J. S. M. P. E. 

The analysis of photographic processing solutions may be of great 
assistance to the laboratory chemist from several standpoints such as : 

(1) A complete chemical check-up of all solutions prepared by him prior to 
their actual production use. 

(2) A definite aid in maintaining uniformity of photographic results through 
properly maintained solutions. 

(3) A check on solutions having unusual chemical behavior. 

(4) A direct saving from the economic standpoint through more efficient use of 
chemicals and elimination of undue waste. 

Previous Work on the Chemistry of Processing Solutions. The 
analysis of organic developing agents has been described by H. T. 
Clarke 1 in his work, "The Examination of Organic Developing 
Agents," Ermen 2 in "Qualitative Tests for the Commoner Develop- 
ers," and Plaumann 3 in "Erkennung von Entwicklersubstanzen." 

The original work which was done on the correlation of developer 
activity and chemical constitution throughout the useful life of a de- 
veloper was by Lehmann and Tausch. 4 - 6 

Evans 6 in his paper, "Maintenance of Developer by Continuous 
Replenishment," discusses the application of analytical controls to 
developer maintenance. Two more recent papers by Evans and 
Hanson 7 and Evans and Silberstein 8 discuss practical methods of 
developer analysis to be used by motion picture laboratories. 

H. L. Baumbach 9 discussed the determination of elon, hydroqui- 
none, and bromide by volumetric methods in a recent article in THIS 

Standard texts in the field of inorganic chemistry are Prescott and 
Johnson 10 and Treadwell and Hall; 11 for organic analysis Kamm 12 and 
the Association of Official Agricultural Chemists, 13 and in the field of 
microchemistry Benedetti-Pichler-Spikes 14 and Chamot and Mason. 16 



Methods are given for analyzing for the following developing 

Trade Name Chemical Name 

Elon Monomethyl ^-aminophenol sulfate 

Hydroquinone Quinol 

Pyro Pyrogallol 

P-a.p., Kodelon p-Aminophenol oxalate 

Glycine, Athenon -Hydroxyphenyl glycine 

May, 1940] 




2,4-Diaminophenol hydrochloride 
/>-Phenylenediamine hydrochloride 

NOTE : These compounds will be referred to in this paper by the names com- 
monly used in the trade. 


When dealing with used developers, it is necessary to distinguish 
between the original developing agent and its oxidized form. In the 










FIG. 1. Z7-tube extractor. 

specific cases of elon and hydroquinone, it has been fairly well estab- 
lished that they are oxidized to their respective sulfonates, and these 
are formed when oxidation is effected by either oxygen or silver 
bromide. Careful work has indicated that these sulfonates are them- 
selves developing agents but that they may be safely neglected in 
first-order approximations of the developer activity. 

The sulfonates are more soluble in water than the original develop- 
ing agents, while the reverse is true in organic solvents, so that elon 
and hydroquinone may be separated from their sulfonates by ex- 
tracting them from an aqueous solution with ethyl ether or ethyl 

488 R. B. ATKINSON AND V. C. SHANER [J. S. M. P. E. 

acetate. No other suitable method has been found, so extraction of 
the unoxidized developing agents becomes the first step in any de- 
veloper analysis. 

A separatory funnel manually operated is adequate for qualitative 
tests. This may also be suitable for control purposes if the manual 
extraction is carried out under standardized conditions. This may 
not, however, give sufficient precision for quantitative methods, and a 
special extractor has been found preferable. 

The extractor is illustrated in Fig. 1. It consists of a 150-ml 
cylindrical glass funnel with stopcock, connected by rubber tubing to 
a glass U-tube. The Z7-tube is blown with a capillary at the bottom, 
which provides for the formation of small bubbles of the extracting 
solvent and thus insures efficient extraction. One end of the U is bent 
to allow the solvent to run into a receptacle. The dimensions of the 
tube are important, and those given in Fig. 1 have proved satisfactory 
for the analysis of 5.0-ml samples of developer. This size sample is 
suitable for the average motion picture developer, but for abnormal 
developers the dimensions can of course be altered. 

When the extraction method is used, developing agents divide 
themselves into two classes. The first class contains the agents 
which are free organic compounds, typified by hydroquinone, and the 
second class contains those agents which are the salts of an organic 
base and a strong acid, typified by elon. The distinguishing property 
between these classes is that the former may be extracted from any 
aqueous solution by an organic solvent such as ethyl ether or ethyl 
acetate. The second class, however, is made up of agents which are 
themselves insoluble in these solvents, but whose organic bases are 
soluble. In order to extract agents of the second class from aqueous 
solution it is necessary that the free base be present in the solution. 
The maximum concentration of elon base occurs at a pH value of 
approximately 8.0. Extraction with a solvent then removes from 
solution the organic base, and it is possible to determine the weight, 
or other property, of the free base and thus determine quantitatively 
the amount of developing agent originally present in the solution. 


Hydroquinone, also known as quinol, or 1,4-dihydroxy benzene, is 
widely used as a developing agent, generally in combination with elon. 
In appearance it is white, usually crystallized in thin needles, readily 
soluble in water and ethyl ether. 


Although hydroquinone is commonly used with elon, it differs 
from it photographically. In carbonate developers, hydroquinone is 
a powerful developer producing images of great contrast. Its activity 
drops rapidly with decreasing pH, and hydroquinone alone will not 
develop in a borax developer. In combination with elon, however, 
hydroquinone does serve as a reducing agent even in a borax de- 

Qualitative Tests for Hydroquinone. Hydroquinone may be dis- 
tinguished from elon and other salts of organic bases by the fact that 
it is readily soluble in ether, while the latter are not. The addition of 
ferric chloride solution to a concentrated solution of hydroquinone 
will cause it to turn dark brown (elon solution turns purple) and 
quinone is formed. This may be detected by the pungent odor given 
off on heating. Hydroquinone crystals have a definite melting point 
at 169C, and this may be used as a positive means of identification. 

Quantitative Determination of Hydroquinone. The determination 
of hydroquinone in a developer is performed with the aid of the U- 
tube extractor described above. 

A suitable sample (5.0 ml) is pipetted into the funnel of the ex- 
tractor, with the stopcock closed. One drop of a 0.04 per cent solution 
of thymol blue indicator is added, and concentrated hydrochloric acid 
added drop by drop until the color has changed from yellow to red. 

At this point the pH value is approximately 2, where hydroquinone 
is soluble in ether but elon is present in an insoluble form. Enough 
dry sodium chloride is added to saturate the solution, plus a slight 
excess. This decreases the solubility of the hydroquinone in the 
water, and the excess aids in obtaining the desired agitation during 
extraction. The stopcock is then opened, and ethyl ether added 
through the funnel. It is desirable to use C. P. rather than U. S. P. 
ether. The ether flow is regulated by the stopcock, and is allowed to 
run just fast enough so that the water layer is not carried out of the 
tube. When 100 ml of ether have been collected (a graduate is a 
suitable receptacle), water is added through the funnel and the 
ether remaining in the tube is carried over, without carrying over any 

A heavy-walled 500-ml flask is used for the titration. In it are 
placed 200 ml of hot water and 1.0 ml of concentrated hydrochloric 
acid. The ether is added to the flask, which is then connected to an 
aspirator and the ether removed by vacuum. The hydroquinone is 
dissolved by the water. If an aspirator is not available, the ether 

490 R. B. ATKINSON AND V. C. SHANER (j. s. M. P. E. 

may be driven off by heating the flask, taking care not to ignite the 
ether vapor. The solution must be cooled to room temperature be- 
fore titrating. 

When the ether is removed, add 2 ml starch solution and titrate 
with 0.01 normal iodine. This serves to remove any sulfur dioxide 
which is present, and also establishes the blank for the titration. 
The value is not recorded. Because the solution is acid, the hydro- 
quinone will not react with the iodine. 

A buffer solution is now used, containing 30 grams of crystalline 
disodium phosphate and 10 grams of Kodalk per liter. One hundred 
ml of this buffer solution are added to the flask after the first iodine 
titration. This raises the pH of the solution to approximately 7.5, 
where both elon and hydroquinone will react with iodine. The 
blue color will be bleached by the hydroquinone which is now reac- 
tive, and a second titration is carried out, this time recording the 

The titration should be carried just far enough to give a permanent 
blue color, which will not fade in at least one minute. The con- 
centration of hydroquinone in the developer may then be calculated 
from the following equation: 

volume of iodine X normality of iodine X 55 = gramg of hydroquinone ^ liter 
volume of sample o f sample 

If the normality of iodine used